Data Labeling Operations Platform Market Outlook

According to our latest research, the global Data Labeling Operations Platform market size reached USD 2.4 billion in 2024, reflecting the sector's rapid adoption across various industries. The market is expected to grow at a robust CAGR of 23.7% from 2025 to 2033, propelling the market to an estimated USD 18.3 billion by 2033. This remarkable growth trajectory is underpinned by the surging demand for high-quality labeled data to power artificial intelligence (AI) and machine learning (ML) applications, which are becoming increasingly integral to digital transformation strategies across sectors.

The primary growth driver for the Data Labeling Operations Platform market is the exponential rise in AI and ML adoption across industries such as healthcare, automotive, BFSI, and retail. As organizations seek to enhance automation, predictive analytics, and customer experiences, the need for accurately labeled datasets has become paramount. Data labeling platforms are pivotal in streamlining annotation workflows, reducing manual errors, and ensuring consistency in training datasets. This, in turn, accelerates the deployment of AI-powered solutions, creating a virtuous cycle of investment and innovation in data labeling technologies. Furthermore, the proliferation of unstructured data, especially from IoT devices, social media, and enterprise systems, has intensified the need for scalable and efficient data labeling operations, further fueling market expansion.

Another significant factor contributing to market growth is the evolution of data privacy regulations and ethical AI mandates. Enterprises are increasingly prioritizing data governance and transparent AI development, which necessitates robust data labeling operations that can provide audit trails and compliance documentation. Data labeling platforms are now integrating advanced features such as workflow automation, quality assurance, and secure data handling to address these regulatory requirements. This has led to increased adoption among highly regulated industries such as healthcare and finance, where the stakes for data accuracy and compliance are exceptionally high. Additionally, the rise of hybrid and remote work models has prompted organizations to seek cloud-based data labeling solutions that enable seamless collaboration and scalability, further boosting the market.

The market's growth is also propelled by advancements in automation technologies within data labeling platforms. The integration of AI-assisted annotation tools, active learning, and human-in-the-loop frameworks has significantly improved the efficiency and accuracy of data labeling processes. These innovations reduce the dependency on manual labor, lower operational costs, and accelerate project timelines, making data labeling more accessible to organizations of all sizes. As a result, small and medium enterprises (SMEs) are increasingly investing in data labeling operations platforms to gain a competitive edge through AI-driven insights. The continuous evolution of data labeling tools to support new data types, languages, and industry-specific requirements ensures sustained market momentum.

Cloud Labeling Software has emerged as a pivotal solution in the data labeling operations platform market, offering unparalleled scalability and flexibility. As organizations increasingly adopt cloud-based solutions, Cloud Labeling Software enables seamless integration with existing IT infrastructures, allowing for efficient data management and processing. This software is particularly beneficial for enterprises with geographically dispersed teams, as it supports real-time collaboration and centralized project oversight. Furthermore, the cloud-based approach reduces the need for significant upfront investments in hardware, making it an attractive option for businesses of all sizes. The ability to scale operations quickly and efficiently in response to fluctuating workloads is a key advantage, driving the adoption of Cloud Labeling Software across various industries.

Regionally, North America continues to dominate the Data Labeling Operations Platform market, driven by a mature AI ecosystem, substantial technology investments, and a strong presence of leading platform providers. However, the Asia Pacific region is emerging as a high-growth mar

Description: Downsized (256x256) camera trap images used for the analyses in "Can CNN-based species classification generalise across variation in habitat within a camera trap survey?", and the dataset composition for each analysis. Note that images tagged as 'human' have been removed from this dataset. Full-size images for the BorneoCam dataset will be made available at LILA.science. The full SAFE camera trap dataset metadata is available at DOI: 10.5281/zenodo.6627707. Project: This dataset was collected as part of the following SAFE research project: Machine learning and image recognition to monitor spatio-temporal changes in the behaviour and dynamics of species interactions Funding: These data were collected as part of research funded by:

NERC (NERC QMEE CDT Studentship, NE/P012345/1, http://gotw.nerc.ac.uk/list_full.asp?pcode=NE%2FP012345%2F1&cookieConsent=A) This dataset is released under the CC-BY 4.0 licence, requiring that you cite the dataset in any outputs, but has the additional condition that you acknowledge the contribution of these funders in any outputs.

XML metadata: GEMINI compliant metadata for this dataset is available here Files: This dataset consists of 3 files: CT_image_data_info2.xlsx, DN_256x256_image_files.zip, DN_generalisability_code.zip CT_image_data_info2.xlsx This file contains dataset metadata and 1 data tables:

Dataset Images (described in worksheet Dataset_images) Description: This worksheet details the composition of each dataset used in the analyses Number of fields: 69 Number of data rows: 270287 Fields:

filename: Root ID (Field type: id) camera_trap_site: Site ID for the camera trap location (Field type: location) taxon: Taxon recorded by camera trap (Field type: taxa) dist_level: Level of disturbance at site (Field type: ordered categorical) baseline: Label as to whether image is included in the baseline training, validation (val) or test set, or not included (NA) (Field type: categorical) increased_cap: Label as to whether image is included in the 'increased cap' training, validation (val) or test set, or not included (NA) (Field type: categorical) dist_individ_event_level: Label as to whether image is included in the 'individual disturbance level datasets split at event level' training, validation (val) or test set, or not included (NA) (Field type: categorical) dist_combined_event_level_1: Label as to whether image is included in the 'disturbance level combination analysis split at event level: disturbance level 1' training or test set, or not included (NA) (Field type: categorical) dist_combined_event_level_2: Label as to whether image is included in the 'disturbance level combination analysis split at event level: disturbance level 2' training or test set, or not included (NA) (Field type: categorical) dist_combined_event_level_3: Label as to whether image is included in the 'disturbance level combination analysis split at event level: disturbance level 3' training or test set, or not included (NA) (Field type: categorical) dist_combined_event_level_4: Label as to whether image is included in the 'disturbance level combination analysis split at event level: disturbance level 4' training or test set, or not included (NA) (Field type: categorical) dist_combined_event_level_5: Label as to whether image is included in the 'disturbance level combination analysis split at event level: disturbance level 5' training or test set, or not included (NA) (Field type: categorical) dist_combined_event_level_pair_1_2: Label as to whether image is included in the 'disturbance level combination analysis split at event level: disturbance levels 1 and 2 (pair)' training set, or not included (NA) (Field type: categorical) dist_combined_event_level_pair_1_3: Label as to whether image is included in the 'disturbance level combination analysis split at event level: disturbance levels 1 and 3 (pair)' training set, or not included (NA) (Field type: categorical) dist_combined_event_level_pair_1_4: Label as to whether image is included in the 'disturbance level combination analysis split at event level: disturbance levels 1 and 4 (pair)' training set, or not included (NA) (Field type: categorical) dist_combined_event_level_pair_1_5: Label as to whether image is included in the 'disturbance level combination analysis split at event level: disturbance levels 1 and 5 (pair)' training set, or not included (NA) (Field type: categorical) dist_combined_event_level_pair_2_3: Label as to whether image is included in the 'disturbance level combination analysis split at event level: disturbance levels 2 and 3 (pair)' training set, or not included (NA) (Field type: categorical) dist_combined_event_level_pair_2_4: Label as to whether image is included in the 'disturbance level combination analysis split at event level: disturbance levels 2 and 4 (pair)' training set, or not included (NA) (Field type: categorical) dist_combined_event_level_pair_2_5: Label as to whether image is included in the 'disturbance level combination analysis split at event level: disturbance levels 2 and 5 (pair)' training set, or not included (NA) (Field type: categorical) dist_combined_event_level_pair_3_4: Label as to whether image is included in the 'disturbance level combination analysis split at event level: disturbance levels 3 and 4 (pair)' training set, or not included (NA) (Field type: categorical) dist_combined_event_level_pair_3_5: Label as to whether image is included in the 'disturbance level combination analysis split at event level: disturbance levels 3 and 5 (pair)' training set, or not included (NA) (Field type: categorical) dist_combined_event_level_pair_4_5: Label as to whether image is included in the 'disturbance level combination analysis split at event level: disturbance levels 4 and 5 (pair)' training set, or not included (NA) (Field type: categorical) dist_combined_event_level_triple_1_2_3: Label as to whether image is included in the 'disturbance level combination analysis split at event level: disturbance levels 1, 2 and 3 (triple)' training set, or not included (NA) (Field type: categorical) dist_combined_event_level_triple_1_2_4: Label as to whether image is included in the 'disturbance level combination analysis split at event level: disturbance levels 1, 2 and 4 (triple)' training set, or not included (NA) (Field type: categorical) dist_combined_event_level_triple_1_2_5: Label as to whether image is included in the 'disturbance level combination analysis split at event level: disturbance levels 1, 2 and 5 (triple)' training set, or not included (NA) (Field type: categorical) dist_combined_event_level_triple_1_3_4: Label as to whether image is included in the 'disturbance level combination analysis split at event level: disturbance levels 1, 3 and 4 (triple)' training set, or not included (NA) (Field type: categorical) dist_combined_event_level_triple_1_3_5: Label as to whether image is included in the 'disturbance level combination analysis split at event level: disturbance levels 1, 3 and 5 (triple)' training set, or not included (NA) (Field type: categorical) dist_combined_event_level_triple_1_4_5: Label as to whether image is included in the 'disturbance level combination analysis split at event level: disturbance levels 1, 4 and 5 (triple)' training set, or not included (NA) (Field type: categorical) dist_combined_event_level_triple_2_3_4: Label as to whether image is included in the 'disturbance level combination analysis split at event level: disturbance levels 2, 3 and 4 (triple)' training set, or not included (NA) (Field type: categorical) dist_combined_event_level_triple_2_3_5: Label as to whether image is included in the 'disturbance level combination analysis split at event level: disturbance levels 2, 3 and 5 (triple)' training set, or not included (NA) (Field type: categorical) dist_combined_event_level_triple_2_4_5: Label as to whether image is included in the 'disturbance level combination analysis split at event level: disturbance levels 2, 4 and 5 (triple)' training set, or not included (NA) (Field type: categorical) dist_combined_event_level_triple_3_4_5: Label as to whether image is included in the 'disturbance level combination analysis split at event level: disturbance levels 3, 4 and 5 (triple)' training set, or not included (NA) (Field type: categorical) dist_combined_event_level_quad_1_2_3_4: Label as to whether image is included in the 'disturbance level combination analysis split at event level: disturbance levels 1, 2, 3 and 4 (quad)' training set, or not included (NA) (Field type: categorical) dist_combined_event_level_quad_1_2_3_5: Label as to whether image is included in the 'disturbance level combination analysis split at event level: disturbance levels 1, 2, 3 and 5 (quad)' training set, or not included (NA) (Field type: categorical) dist_combined_event_level_quad_1_2_4_5: Label as to whether image is included in the 'disturbance level combination analysis split at event level: disturbance levels 1, 2, 4 and 5 (quad)' training set, or not included (NA) (Field type: categorical) dist_combined_event_level_quad_1_3_4_5: Label as to whether image is included in the 'disturbance level combination analysis split at event level: disturbance levels 1, 3, 4 and 5 (quad)' training set, or not included (NA) (Field type: categorical) dist_combined_event_level_quad_2_3_4_5: Label as to whether image is included in the 'disturbance level combination analysis split at event level: disturbance levels 2, 3, 4 and 5 (quad)' training set, or not included (NA) (Field type: categorical) dist_combined_event_level_all_1_2_3_4_5: Label as to whether image is included in the 'disturbance level combination analysis split at event level: disturbance levels 1, 2, 3, 4 and 5 (all)' training set, or not included (NA) (Field type: categorical) dist_camera_level_individ_1: Label as to whether image is included in the 'disturbance level combination analysis split at camera level: disturbance

Handwritten Digits Pixel Dataset - Documentation

Overview

The Handwritten Digits Pixel Dataset is a collection of numerical data representing handwritten digits from 0 to 9. Unlike image datasets that store actual image files, this dataset contains pixel intensity values arranged in a structured tabular format, making it ideal for machine learning and data analysis applications.

Dataset Description

Basic Information

• Format: CSV (Comma-Separated Values)
• Total Samples: [Number of rows based on your dataset]
• Features: 784 pixel columns (28×28 pixels) + 1 label column
• Label Range: Digits 0-9
• Pixel Value Range: 0-255 (grayscale intensity)

File Structure

Column Description

• label: The target variable representing the digit (0-9)
• pixel columns: 784 columns named in format [row]xcolumn
• Each pixel column contains integer values from 0-255 representing grayscale intensity

Data Characteristics

Label Distribution

The dataset contains handwritten digit samples with the following distribution:

• Digit 0: [X] samples
• Digit 1: [X] samples
• Digit 2: [X] samples
• Digit 3: [X] samples
• Digit 4: [X] samples
• Digit 5: [X] samples
• Digit 6: [X] samples
• Digit 7: [X] samples
• Digit 8: [X] samples
• Digit 9: [X] samples

(Note: Actual distribution counts would be calculated from your specific dataset)

Data Quality

• Missing Values: No missing values detected
• Data Type: All values are integers
• Normalization: Pixel values range from 0-255 (can be normalized to 0-1 for ML models)
• Consistency: Uniform 28×28 grid structure across all samples

Technical Specifications

Data Preprocessing Requirements

• Normalization: Scale pixel values from 0-255 to 0-1 range
• Reshaping: Convert 1D pixel arrays to 2D 28×28 matrices for visualization
• Train-Test Split: Recommended 80-20 or 70-30 split for model development

Recommended Machine Learning Approaches

Classification Algorithms:

• Random Forest
• Support Vector Machines (SVM)
• Neural Networks
• K-Nearest Neighbors (KNN)

Deep Learning Architectures:

• Convolutional Neural Networks (CNNs)
• Multi-layer Perceptrons (MLPs)

Dimensionality Reduction:

• PCA (Principal Component Analysis)
• t-SNE for visualization

Usage Examples

Loading the Dataset

import pandas as pd

# Load the dataset
df = pd.read_csv('/kaggle/input/handwritten_digits_pixel_dataset/mnist.csv')

# Separate features and labels
X = df.drop('label', axis=1)
y = df['label']

# Normalize pixel values
X_normalized = X / 255.0

This is Part 2/2 of the ActiveHuman dataset! Part 1 can be found here. Dataset Description ActiveHuman was generated using Unity's Perception package. It consists of 175428 RGB images and their semantic segmentation counterparts taken at different environments, lighting conditions, camera distances and angles. In total, the dataset contains images for 8 environments, 33 humans, 4 lighting conditions, 7 camera distances (1m-4m) and 36 camera angles (0-360 at 10-degree intervals). The dataset does not include images at every single combination of available camera distances and angles, since for some values the camera would collide with another object or go outside the confines of an environment. As a result, some combinations of camera distances and angles do not exist in the dataset. Alongside each image, 2D Bounding Box, 3D Bounding Box and Keypoint ground truth annotations are also generated via the use of Labelers and are stored as a JSON-based dataset. These Labelers are scripts that are responsible for capturing ground truth annotations for each captured image or frame. Keypoint annotations follow the COCO format defined by the COCO keypoint annotation template offered in the perception package.

Folder configuration The dataset consists of 3 folders:

JSON Data: Contains all the generated JSON files. RGB Images: Contains the generated RGB images. Semantic Segmentation Images: Contains the generated semantic segmentation images.

Essential Terminology

Annotation: Recorded data describing a single capture. Capture: One completed rendering process of a Unity sensor which stored the rendered result to data files (e.g. PNG, JPG, etc.). Ego: Object or person on which a collection of sensors is attached to (e.g., if a drone has a camera attached to it, the drone would be the ego and the camera would be the sensor). Ego coordinate system: Coordinates with respect to the ego. Global coordinate system: Coordinates with respect to the global origin in Unity. Sensor: Device that captures the dataset (in this instance the sensor is a camera). Sensor coordinate system: Coordinates with respect to the sensor. Sequence: Time-ordered series of captures. This is very useful for video capture where the time-order relationship of two captures is vital. UIID: Universal Unique Identifier. It is a unique hexadecimal identifier that can represent an individual instance of a capture, ego, sensor, annotation, labeled object or keypoint, or keypoint template.

Dataset Data The dataset includes 4 types of JSON annotation files files:

annotation_definitions.json: Contains annotation definitions for all of the active Labelers of the simulation stored in an array. Each entry consists of a collection of key-value pairs which describe a particular type of annotation and contain information about that specific annotation describing how its data should be mapped back to labels or objects in the scene. Each entry contains the following key-value pairs:

id: Integer identifier of the annotation's definition. name: Annotation name (e.g., keypoints, bounding box, bounding box 3D, semantic segmentation). description: Description of the annotation's specifications. format: Format of the file containing the annotation specifications (e.g., json, PNG). spec: Format-specific specifications for the annotation values generated by each Labeler.

Most Labelers generate different annotation specifications in the spec key-value pair:

BoundingBox2DLabeler/BoundingBox3DLabeler:

label_id: Integer identifier of a label. label_name: String identifier of a label. KeypointLabeler:

template_id: Keypoint template UUID. template_name: Name of the keypoint template. key_points: Array containing all the joints defined by the keypoint template. This array includes the key-value pairs:

label: Joint label. index: Joint index. color: RGBA values of the keypoint. color_code: Hex color code of the keypoint skeleton: Array containing all the skeleton connections defined by the keypoint template. Each skeleton connection defines a connection between two different joints. This array includes the key-value pairs:

label1: Label of the first joint. label2: Label of the second joint. joint1: Index of the first joint. joint2: Index of the second joint. color: RGBA values of the connection. color_code: Hex color code of the connection. SemanticSegmentationLabeler:

label_name: String identifier of a label. pixel_value: RGBA values of the label. color_code: Hex color code of the label.

captures_xyz.json: Each of these files contains an array of ground truth annotations generated by each active Labeler for each capture separately, as well as extra metadata that describe the state of each active sensor that is present in the scene. Each array entry in the contains the following key-value pairs:

id: UUID of the capture. sequence_id: UUID of the sequence. step: Index of the capture within a sequence. timestamp: Timestamp (in ms) since the beginning of a sequence. sensor: Properties of the sensor. This entry contains a collection with the following key-value pairs:

sensor_id: Sensor UUID. ego_id: Ego UUID. modality: Modality of the sensor (e.g., camera, radar). translation: 3D vector that describes the sensor's position (in meters) with respect to the global coordinate system. rotation: Quaternion variable that describes the sensor's orientation with respect to the ego coordinate system. camera_intrinsic: matrix containing (if it exists) the camera's intrinsic calibration. projection: Projection type used by the camera (e.g., orthographic, perspective). ego: Attributes of the ego. This entry contains a collection with the following key-value pairs:

ego_id: Ego UUID. translation: 3D vector that describes the ego's position (in meters) with respect to the global coordinate system. rotation: Quaternion variable containing the ego's orientation. velocity: 3D vector containing the ego's velocity (in meters per second). acceleration: 3D vector containing the ego's acceleration (in ). format: Format of the file captured by the sensor (e.g., PNG, JPG). annotations: Key-value pair collections, one for each active Labeler. These key-value pairs are as follows:

id: Annotation UUID . annotation_definition: Integer identifier of the annotation's definition. filename: Name of the file generated by the Labeler. This entry is only present for Labelers that generate an image. values: List of key-value pairs containing annotation data for the current Labeler.

Each Labeler generates different annotation specifications in the values key-value pair:

BoundingBox2DLabeler:

label_id: Integer identifier of a label. label_name: String identifier of a label. instance_id: UUID of one instance of an object. Each object with the same label that is visible on the same capture has different instance_id values. x: Position of the 2D bounding box on the X axis. y: Position of the 2D bounding box position on the Y axis. width: Width of the 2D bounding box. height: Height of the 2D bounding box. BoundingBox3DLabeler:

label_id: Integer identifier of a label. label_name: String identifier of a label. instance_id: UUID of one instance of an object. Each object with the same label that is visible on the same capture has different instance_id values. translation: 3D vector containing the location of the center of the 3D bounding box with respect to the sensor coordinate system (in meters). size: 3D vector containing the size of the 3D bounding box (in meters) rotation: Quaternion variable containing the orientation of the 3D bounding box. velocity: 3D vector containing the velocity of the 3D bounding box (in meters per second). acceleration: 3D vector containing the acceleration of the 3D bounding box acceleration (in ). KeypointLabeler:

label_id: Integer identifier of a label. instance_id: UUID of one instance of a joint. Keypoints with the same joint label that are visible on the same capture have different instance_id values. template_id: UUID of the keypoint template. pose: Pose label for that particular capture. keypoints: Array containing the properties of each keypoint. Each keypoint that exists in the keypoint template file is one element of the array. Each entry's contents have as follows:

index: Index of the keypoint in the keypoint template file. x: Pixel coordinates of the keypoint on the X axis. y: Pixel coordinates of the keypoint on the Y axis. state: State of the keypoint.

The SemanticSegmentationLabeler does not contain a values list.

egos.json: Contains collections of key-value pairs for each ego. These include:

id: UUID of the ego. description: Description of the ego. sensors.json: Contains collections of key-value pairs for all sensors of the simulation. These include:

id: UUID of the sensor. ego_id: UUID of the ego on which the sensor is attached. modality: Modality of the sensor (e.g., camera, radar, sonar). description: Description of the sensor (e.g., camera, radar).

Image names The RGB and semantic segmentation images share the same image naming convention. However, the semantic segmentation images also contain the string Semantic_ at the beginning of their filenames. Each RGB image is named "e_h_l_d_r.jpg", where:

e denotes the id of the environment. h denotes the id of the person. l denotes the id of the lighting condition. d denotes the camera distance at which the image was captured. r denotes the camera angle at which the image was captured.

General Information

This dataset was created for the training and testing of machine learning systems for extracting information from slates/on-screen or filmed text in video productions. The data associated with each instance was acquired by observing text on the slates in the file. There are two levels of data collected, a direct transcription and contextual information. For the direct transcription if there was illegible text an approximation was derived. The information is reported by the original creator of the slates and can be assumed to be accurate.

The data was collected using a software made specifically to categorize and transcribe metadata from these instances (see file directory description). The transcription was written in a natural reading order (for a western audience), so right to left and top to bottom. If the instance was labeled “Graphical” then the reading order was also right to left and top to bottom within individual sections as well as work as a whole.

This dataset was created by Madison Courtney, in collaboration with GBH Archives staff, and in consultation with researchers in the Brandeis University Department of Computer Science.

Uniqueness and overlapping data

Some of the slates come from different episodes of the same series; therefore, some slates have data overlap. For example, the “series-title” may be common across many slates. However, each slate instance in this dataset was labeled independently of the others. No information was removed, but not every slate contains the same information.

Different “sub-types” of slates have different graphical features, and present unique challenges for interpretation. In general, sub-types H (Handwritten), G (Graphical), C (Clapperboard) are more complex than D (Simple digital text) and B (Slate over bars). Most instances in the dataset are D. Users may wish to restrict the set to only those with subtype D.

Labels and annotations were created by an expert human judge. In Version 2, labels and annotations were created only once without any measure of inter-annotator agreement. In Version 3, all data were confirmed and/or edited by a second expert human judge. The dataset is self-contained. But more information about the assets from which these slates were taken can be found at the main website of the AAPB https://www.americanarchive.org/

Data size and structure

The data is tabular. There are 7 columns and 503 rows. Each row represents a different labeled image. The image files themselves are included in the dataset directory. The columns are as follows:

• 0: filename : The name of the image file for this slate
• 1: seen : A boolean book-keeping field used during the annotation process
• 2: type-label : The type of scene pictured in the image. All images in this set have type "S" signifying "Slate"
• 3: subtype-label : The sub-type of scene pictured in the image. Possible subtypes are "H" (Handwritten), "C" (Clapperboard), "D" (Simple digital text), "B" (Slate over bars), "G" (Graphical).
• 4: modifier : A boolean value indicating whether the slate was "transitional" in the sense that the still image was captured as the slate was fading in or out of view.
• 5: note-3 : Verbatim transcription of the text appearing on the slate
• 6: note-4 : Data in key-value structure indicating important data values presented on the slate. Possible keys are "program-title", "episode-title", "series-title", "title", "episode-no", "create-date", "air-date", "date", "director", "producer", "camera". Dates were normalized as YYYY-MM-DD. Names were normalized as Last, First Middle.

Data format

The directory contains the tabular data, the image files, and a small utility for viewing and/or editing labels. The Keystroke Labeler utility is a simple, serverless HTML-based viewer/editor. You can use the Keystroke Labeler by simply opening labeler.html in your web browser. The data are also provided serialized as JSON and CSV. The exact same label data appears redundantly in these 3 files: - img_arr_prog.js - the label data loaded by the Keystroke Labeler - img_labels.csv - the label data serialized as CSV - img_labels.json - the label data serialized as JSON

This dataset includes metadata about programs in the American Archive of Public Broadcasting. Any use of programs referenced by this dataset are subject to the terms of use set by the American Archive of Public Broadcasting.

Data Label Quality Assurance for AVs Market Outlook

According to our latest research, the global Data Label Quality Assurance for AVs market size reached USD 1.12 billion in 2024, with a robust compound annual growth rate (CAGR) of 13.8% projected through the forecast period. By 2033, the market is expected to achieve a value of USD 3.48 billion, highlighting the increasing importance of high-quality data annotation and verification in the autonomous vehicle (AV) ecosystem. This growth is primarily driven by the surging adoption of advanced driver-assistance systems (ADAS), rapid advancements in sensor technologies, and the critical need for precise, reliable labeled data to train and validate machine learning models powering AVs.

The exponential growth factor for the Data Label Quality Assurance for AVs market is rooted in the escalating complexity and data requirements of autonomous driving systems. As AVs rely heavily on artificial intelligence and machine learning algorithms, the accuracy of labeled data directly impacts safety, efficiency, and performance. The proliferation of multi-sensor fusion technologies, such as LiDAR, radar, and high-definition cameras, has resulted in massive volumes of heterogeneous data streams. Ensuring the quality and consistency of labeled datasets, therefore, becomes indispensable for reducing algorithmic bias, minimizing false positives, and enhancing real-world deployment reliability. Furthermore, stringent regulatory frameworks and safety standards enforced by governments and industry bodies have amplified the demand for comprehensive quality assurance protocols in data labeling workflows, making this market a central pillar in the AV development lifecycle.

Another significant driver is the expanding ecosystem of industry stakeholders, including OEMs, Tier 1 suppliers, and technology providers, all of whom are investing heavily in AV R&D. The competitive race to commercialize Level 4 and Level 5 autonomous vehicles has intensified the focus on data integrity, encouraging the adoption of advanced QA solutions that combine manual expertise with automated validation tools. Additionally, the growing trend towards hybrid QA approaches—integrating human-in-the-loop verification with AI-powered quality checks—enables higher throughput and scalability without compromising annotation accuracy. This evolution is further supported by the rise of cloud-based platforms and collaborative tools, which facilitate seamless data sharing, version control, and cross-functional QA processes across geographically dispersed teams.

On the regional front, North America continues to lead the Data Label Quality Assurance for AVs market, propelled by the presence of major automotive innovators, tech giants, and a mature regulatory environment conducive to AV testing and deployment. The Asia Pacific region, meanwhile, is emerging as a high-growth market, driven by rapid urbanization, government-backed smart mobility initiatives, and the burgeoning presence of local technology providers specializing in data annotation services. Europe also maintains a strong foothold, benefiting from a robust automotive sector, cross-border R&D collaborations, and harmonized safety standards. These regional dynamics collectively shape a highly competitive and innovation-driven global market landscape.

Solution Type Analysis

The Solution Type segment of the Data Label Quality Assurance for AVs market encompasses Manual QA, Automated QA, and Hybrid QA. Manual QA remains a foundational approach, particularly for complex annotation tasks that demand nuanced human judgment and domain expertise. This method involves skilled annotators meticulously reviewing and validating labeled datasets to ensure compliance with predefined quality metrics. While manual QA is resource-intensive and time-consuming, it is indispensable for tasks requiring contextual understanding, such as semantic segmentation and rare object identification. The continued reliance on manual QA is also driven by the need to address edge cases and ambiguous scenarios that autom

The AI data labeling services market is experiencing robust growth, driven by the increasing adoption of artificial intelligence across various sectors. The market's expansion is fueled by the critical need for high-quality labeled data to train and improve the accuracy of AI algorithms. While precise figures for market size and CAGR are not provided, industry reports suggest a significant market value, potentially exceeding $5 billion by 2025, with a Compound Annual Growth Rate (CAGR) likely in the range of 25-30% from 2025-2033. This rapid growth is attributed to several factors, including the proliferation of AI applications in autonomous vehicles, healthcare diagnostics, e-commerce personalization, and precision agriculture. The increasing availability of cloud-based solutions is also contributing to market expansion, offering scalability and cost-effectiveness for businesses of all sizes. However, challenges remain, such as the high cost of data annotation, the need for skilled labor, and concerns around data privacy and security. The market is segmented by application (automotive, healthcare, retail, agriculture, others) and type (cloud-based, on-premises), with the cloud-based segment expected to dominate due to its flexibility and accessibility. Key players like Scale AI, Labelbox, and Appen are driving innovation and market consolidation through technological advancements and strategic acquisitions. Geographic growth is expected across all regions, with North America and Asia-Pacific anticipated to lead in market share due to high AI adoption rates and significant investments in technological infrastructure. The competitive landscape is dynamic, featuring both established players and emerging startups. Strategic partnerships and mergers and acquisitions are common strategies for market expansion and technological enhancement. Future growth hinges on advancements in automation technologies that reduce the cost and time associated with data labeling. Furthermore, the development of more robust and standardized quality control metrics will be crucial for assuring the accuracy and reliability of labeled datasets, which is crucial for building trust and furthering adoption of AI-powered applications. The focus on addressing ethical considerations around data bias and privacy will also play a critical role in shaping the market's future trajectory. Continued innovation in both the technology and business models within the AI data labeling services sector will be vital for sustaining the high growth projected for the coming decade.

Data Labeling Services Market Outlook

According to our latest research, the global data labeling services market size reached USD 2.5 billion in 2024, reflecting robust demand across multiple industries driven by the rapid proliferation of artificial intelligence (AI) and machine learning (ML) applications. The market is anticipated to grow at a CAGR of 22.1% from 2025 to 2033, with the forecasted market size expected to reach USD 18.6 billion by 2033. This remarkable expansion is primarily attributed to the increasing adoption of AI-powered solutions, the surge in data-driven decision-making, and the ongoing digital transformation across sectors. As per the latest research, key growth factors include the need for high-quality annotated data, the expansion of autonomous technologies, and the rising demand for automation in business processes.

One of the main growth factors accelerating the data labeling services market is the exponential increase in the volume of unstructured data generated daily by enterprises, devices, and consumers. Organizations are seeking advanced AI and ML models to extract actionable insights from this vast data pool. However, the effectiveness of these models is directly linked to the accuracy and quality of labeled data. As a result, businesses are increasingly outsourcing data annotation to specialized service providers, ensuring high accuracy and consistency in labeling tasks. The emergence of sectors such as autonomous vehicles, healthcare diagnostics, and smart retail has further amplified the need for scalable, reliable, and cost-effective data labeling services. Additionally, the proliferation of edge computing and IoT devices is generating diverse data types that require precise annotation, thus fueling market growth.

Another significant driver is the advancement in AI technologies, particularly in computer vision, natural language processing, and speech recognition. The evolution of deep learning algorithms has heightened the demand for comprehensive datasets with meticulous labeling, as these models require vast quantities of annotated images, videos, text, and audio for effective training and validation. This has led to the emergence of new business models in the data labeling ecosystem, including crowd-sourced labeling, managed labeling services, and automated annotation tools. Furthermore, regulatory mandates in sectors like healthcare and automotive, which necessitate the use of ethically sourced and accurately labeled data, are propelling the adoption of professional data labeling services. The increased focus on data privacy and compliance is also prompting organizations to partner with established service providers that adhere to stringent data security protocols.

The integration of data labeling services with advanced technologies such as active learning, human-in-the-loop (HITL) systems, and AI-assisted annotation platforms is further boosting market expansion. These innovations are enhancing the efficiency and scalability of labeling processes, enabling the handling of complex datasets across varied formats. The growing trend of hybrid labeling models, combining manual expertise with automation, is optimizing both accuracy and turnaround times. Moreover, the increasing investments from venture capitalists and technology giants in AI startups and data labeling platforms are fostering the development of innovative solutions, thereby strengthening the market ecosystem. As organizations strive for higher model performance and faster deployment cycles, the demand for specialized, domain-specific labeling services continues to surge.

From a regional perspective, North America remains the dominant market for data labeling services, owing to its strong presence of leading AI technology companies, robust digital infrastructure, and early adoption of advanced analytics. However, Asia Pacific is rapidly emerging as the fastest-growing region, fueled by the expansion of IT outsourcing hubs, the rise of AI startups, and government initiatives promoting digital transformation. Europe is also witnessing significant growth, driven by stringent data privacy regulations and increased investments in AI research. Meanwhile, Latin America and the Middle East & Africa are gradually catching up, as enterprises in these regions recognize the value of annotated data in enhancing operational efficiency and customer experience. The evolving regulatory landscape and the increasing availability of skilled annotators are expected to further accelerate market growth across all regions.

The OSDG Community Dataset (OSDG-CD) is a public dataset of thousands of text excerpts, which were validated by over 1,400 OSDG Community Platform (OSDG-CP) citizen scientists from over 140 countries, with respect to the Sustainable Development Goals (SDGs).

Dataset Information

In support of the global effort to achieve the Sustainable Development Goals (SDGs), OSDG is realising a series of SDG-labelled text datasets. The OSDG Community Dataset (OSDG-CD) is the direct result of the work of more than 1,400 volunteers from over 130 countries who have contributed to our understanding of SDGs via the OSDG Community Platform (OSDG-CP). The dataset contains tens of thousands of text excerpts (henceforth: texts) which were validated by the Community volunteers with respect to SDGs. The data can be used to derive insights into the nature of SDGs using either ontology-based or machine learning approaches.

📘 The file contains 43,0210 (+390) text excerpts and a total of 310,328 (+3,733) assigned labels.

To learn more about the project, please visit the OSDG website and the official GitHub page. Explore a detailed overview of the OSDG methodology in our recent paper "OSDG 2.0: a multilingual tool for classifying text data by UN Sustainable Development Goals (SDGs)".

Source Data

The dataset consists of paragraph-length text excerpts derived from publicly available documents, including reports, policy documents and publication abstracts. A significant number of documents (more than 3,000) originate from UN-related sources such as SDG-Pathfinder and SDG Library. These sources often contain documents that already have SDG labels associated with them. Each text is comprised of 3 to 6 sentences and is about 90 words on average.

Methodology

All the texts are evaluated by volunteers on the OSDG-CP. The platform is an ambitious attempt to bring together researchers, subject-matter experts and SDG advocates from all around the world to create a large and accurate source of textual information on the SDGs. The Community volunteers use the platform to participate in labelling exercises where they validate each text's relevance to SDGs based on their background knowledge.

In each exercise, the volunteer is shown a text together with an SDG label associated with it – this usually comes from the source – and asked to either accept or reject the suggested label.

There are 3 types of exercises:

All volunteers start with the mandatory introductory exercise that consists of 10 pre-selected texts. Each volunteer must complete this exercise before they can access 2 other exercise types. Upon completion, the volunteer reviews the exercise by comparing their answers with the answers of the rest of the Community using aggregated statistics we provide, i.e., the share of those who accepted and rejected the suggested SDG label for each of the 10 texts. This helps the volunteer to get a feel for the platform.

SDG-specific exercises where the volunteer validates texts with respect to a single SDG, e.g., SDG 1 No Poverty.

All SDGs exercise where the volunteer validates a random sequence of texts where each text can have any SDG as its associated label.

After finishing the introductory exercise, the volunteer is free to select either SDG-specific or All SDGs exercises. Each exercise, regardless of its type, consists of 100 texts. Once the exercise is finished, the volunteer can either label more texts or exit the platform. Of course, the volunteer can finish the exercise early. All progress is saved and recorded still.

To ensure quality, each text is validated by up to 9 different volunteers and all texts included in the public release of the data have been validated by at least 3 different volunteers.

It is worth keeping in mind that all exercises present the volunteers with a binary decision problem, i.e., either accept or reject a suggested label. The volunteers are never asked to select one or more SDGs that a certain text might relate to. The rationale behind this set-up is that asking a volunteer to select from 17 SDGs is extremely inefficient. Currently, all texts are validated against only one associated SDG label.

Column Description

doi - Digital Object Identifier of the original document

text_id - unique text identifier

text - text excerpt from the document

sdg - the SDG the text is validated against

labels_negative - the number of volunteers who rejected the suggested SDG label

labels_positive - the number of volunteers who accepted the suggested SDG label

agreement - agreement score based on the formula (agreement = \frac{|labels_{positive} - labels_{negative}|}{labels_{positive} + labels_{negative}})

Further Information

Do not hesitate to share with us your outputs, be it a research paper, a machine learning model, a blog post, or just an interesting observation. All queries can be directed to community@osdg.ai.

BOSQUE Test Set: A Dermoscopic Image Dataset from Colombian Patients with Diverse Skin Phototypes Description: The BOSQUE Test Set is a curated dataset of 151 dermoscopic images of pigmented skin lesions, collected from dermatology consultations and outreach campaigns in Bogotá, Colombia. Each image is accompanied by expert-verified metadata including histological diagnosis, patient demographic details, anatomical site, and skin phototype. The dataset is intended to support machine learning research in dermatology with a particular focus on skin tone diversity and fairness in diagnostic algorithms. The dataset was developed under the guidance of Universidad El Bosque, whose name inspired the acronym BOSQUE. It responds to the global underrepresentation of darker skin phototypes in existing dermoscopic image collections such as HAM10000, and aims to improve diagnostic equity through inclusive data curation. Key Features 151 dermoscopic images acquired in real-world clinical settings Captured using polarized light dermatoscopes (DermLite 4 + iPhone) Inclusive population: Sex: 97 Female, 54 Male Age groups: from 0–29 to 90+, categorized into clinically relevant bins Fitzpatrick skin phototypes: ranging from II to VI Type II (fair, burns easily): 11 patients Type III (light brown, mild burns): 94 patients Type IV (moderate brown, rarely burns): 34 patients Type V (dark brown, very rarely burns): 7 patients Type VI (deeply pigmented, never burns): 5 patients Lesion characteristics: Nature: benign or malignant (histopathologically confirmed) Size: categorized as ≤5mm, 6–10mm, 11–20mm, >20mm Evolution time: grouped into <1y, 1y, 2y, 3–4y, 5–9y, and 10y+ categories Anatomical site: head/neck, trunk, limbs, or acral areas Histopathological diagnosis: 7-class ISIC-style labels (akiec, bcc, bkl, df, mel, nv, vasc) Clinical label: melanocytic vs. non-melanocytic (from clinical diagnosis) Clinical context: includes personal history of NMSC and use of photosensitizing drugs Image naming: pseudonymized file names encode diagnosis label and image ID Ethics: all data anonymized and collected under IRB-approved protocol in Colombia Included Files BOSQUE_test_set.zip: Folder containing 151 dermoscopic image files (JPG) BOSQUE_metadata.csv: Metadata for each image, including: Patient sex, age group, skin phototype Anatomical site of the lesion Lesion nature (benign/malignant) Lesion size and evolution time (binned) Histological diagnosis (7-class) Clinical label (melanocytic / non-melanocytic) Use Cases This dataset is intended for: Benchmarking AI models for dermoscopic image classification Fairness analysis across skin tones, sex, and age groups Medical education and clinical training on diverse skin phototypes Comparison against HAM10000 or ISIC datasets in research Ethical Statement All patients provided informed consent for the capture and use of clinical and dermoscopic images, the collection of relevant clinical metadata, and the performance of skin biopsies for diagnostic confirmation. The study protocol was reviewed and approved by the Institutional Ethics Committee at Subred Integrada de Servicios de Salud Norte E.S.E and Universidad El Bosque (Bogotá, Colombia). All data were anonymized in compliance with Colombian health data privacy regulations and international ethical standards (e.g., Declaration of Helsinki). No personally identifiable information is included in the metadata or image files. Access to data was restricted to authorized investigators, and patients were informed about the research and educational use of their anonymized data. Suggested Citation [Author(s)]. (2025). BOSQUE Test Set: A Dermoscopic Image Dataset from Colombian Patients with Diverse Skin Phototypes [Data set]. Harvard Dataverse. https://doi.org/xxxxx

\(\color{#9911ff}{\mathcal{CONTEXT}}\) This data collection was created for exercises in Machine Learning. Images are generated completely artificially using the math parametric functions with three random coefficients. One of them (an integer number) became the "label" for classification, the other two (real numbers) - the "targets" for regression analysis. Different random colors are planned to be a "noise" for predictions. Of course, the data is free for noncommercial and nongovernmental goals.

\(\color{#9911ff}{\mathcal{CONTENT}}\) The process of data building - Synthetic Data 3. All images, labels, and targets are numeric arrays with the same data types and shapes. They are collected here in .h5 files. In every file: - images (float32 => 288x288 pixels, 3 color channels); - labels (int32 => 7 classes); - targets (float32 => 2 coefficients).

\(\color{#9911ff}{\mathcal{ACKNOWLEDGMENTS}}\) Thanks for your attention.

\(\color{#9911ff}{\mathcal{INSPIRATION}}\) Discovering the capabilities of algorithms in the recognition of absolutely synthetic data.

We performed CODEX (co-detection by indexing) multiplexed imaging on four sections of the human colon (ascending, transverse, descending, and sigmoid) using a panel of 47 oligonucleotide-barcoded antibodies. Subsequently images underwent standard CODEX image processing (tile stitching, drift compensation, cycle concatenation, background subtraction, deconvolution, and determination of best focal plane), and single cell segmentation. Output of this process was a dataframe of nearly 130,000 cells with fluorescence values quantified from each marker. We used this dataframe as input to 1 of the 5 normalization techniques of which we compared z, double-log(z), min/max, and arcsinh normalizations to the original unmodified dataset. We used these normalized dataframes as inputs for 4 unsupervised clustering algorithms: k-means, leiden, X-shift euclidian, and X-shift angular.

From the clustering outputs, we then labeled the clusters that resulted for cells observed in the data producing 20 unique cell type labels. We also labeled cell types by hiearchical hand-gating data within cellengine (cellengine.com). We also created another gold standard for comparison by overclustering unormalized data with X-shift angular clustering. Finally, we created one last label as the major cell type call from each cell from all 21 cell type labels in the dataset.

Consequently the dataset has individual cells segmented out in each row. Then there are columns for the X, Y position in pixels in the overall montage image of the dataset. There are also columns to indicate which region the data came from (4 total). The rest are labels generated by all the clustering and normalization techniques used in the manuscript and what were compared to each other. These also were the data that were used for neighborhood analysis for the last figure of the manuscript. These are provided at all four levels of cell type level granularity (from 7 cell types to 35 cell types).

Labeling Tools for Warehouse Vision Models Market Outlook

According to our latest research, the global market size for Labeling Tools for Warehouse Vision Models reached USD 1.21 billion in 2024, with a robust CAGR of 18.7% projected through the forecast period. By 2033, the market is expected to reach USD 5.89 billion, driven by the increasing adoption of AI-powered vision systems in warehouses for automation and efficiency. The market’s growth is primarily fueled by the rapid digital transformation in the logistics and warehousing sectors, where vision models are revolutionizing inventory management, quality control, and automated sorting processes.

One of the most significant growth factors for the Labeling Tools for Warehouse Vision Models Market is the escalating demand for automation across supply chains and distribution centers. As companies strive to enhance operational efficiency and reduce human error, the integration of advanced computer vision models has become essential. These models, however, require vast amounts of accurately labeled data to function optimally. This necessity has led to a surge in demand for sophisticated labeling tools capable of handling diverse data types, such as images, videos, and 3D point clouds. Moreover, the proliferation of e-commerce and omnichannel retailing has put immense pressure on warehouses to process and ship orders faster, further fueling the need for robust labeling solutions that can support rapid model development and deployment.

Another key driver is the evolution of warehouse robotics and autonomous systems. Modern warehouses are increasingly deploying robots and automated guided vehicles (AGVs) that rely on vision models for navigation, object detection, and picking operations. For these systems to perform accurately, high-quality annotated datasets are crucial. The growing complexity and variety of warehouse environments also necessitate labeling tools that can adapt to different use cases, such as detecting damaged goods, monitoring shelf inventory, and facilitating automated sorting. As a result, vendors are innovating their labeling platforms to offer features like collaborative annotation, AI-assisted labeling, and integration with warehouse management systems, all of which are contributing to market growth.

Additionally, the rise of cloud computing and advancements in machine learning infrastructure are accelerating the adoption of labeling tools in the warehouse sector. Cloud-based labeling platforms offer scalability, remote collaboration, and seamless integration with AI training pipelines, making them highly attractive for large enterprises and third-party logistics providers. These solutions enable warehouses to manage vast datasets, ensure data security, and accelerate the development of vision models. Furthermore, regulatory requirements for traceability and quality assurance in industries such as pharmaceuticals and food & beverage are driving warehouses to invest in state-of-the-art vision models, thereby increasing the demand for comprehensive labeling tools.

From a regional perspective, North America currently leads the Labeling Tools for Warehouse Vision Models Market, accounting for the largest market share in 2024. This dominance is attributed to the early adoption of warehouse automation technologies, a strong presence of leading logistics and e-commerce players, and significant investments in AI research and development. The Asia Pacific region is poised for the fastest growth, supported by the rapid expansion of manufacturing and e-commerce sectors in countries like China, India, and Japan. Europe also presents lucrative opportunities due to stringent quality control regulations and growing focus on supply chain digitization. Meanwhile, Latin America and the Middle East & Africa are gradually catching up, driven by increasing investments in logistics infrastructure and digital transformation initiatives.

Product Type Analysis

The Product Type segment of the Labeling Tools for Warehouse Vi

One of the most challenging issues in machine learning is imbalanced data analysis. Usually, in this type of research, correctly predicting minority labels is more critical than correctly predicting majority labels. However, traditional machine learning techniques easily lead to learning bias. Traditional classifiers tend to place all subjects in the majority group, resulting in biased predictions. Machine learning studies are typically conducted from one of two perspectives: a data-based perspective or a model-based perspective. Oversampling and undersampling are examples of data-based approaches, while the addition of costs, penalties, or weights to optimize the algorithm is typical of a model-based approach. Some ensemble methods have been studied recently. These methods cause various problems, such as overfitting, the omission of some information, and long computation times. In addition, these methods do not apply to all kinds of datasets. Based on this problem, the virtual labels (ViLa) approach for the majority label is proposed to solve the imbalanced problem. A new multiclass classification approach with the equal K-means clustering method is demonstrated in the study. The proposed method is compared with commonly used imbalance problem methods, such as sampling methods (oversampling, undersampling, and SMOTE) and classifier methods (SVM and one-class SVM). The results show that the proposed method performs better when the degree of data imbalance increases and will gradually outperform other methods.

Purpose – Introducing Environmental Labels for aircraft according to the ISO 14025 standard allowing to compare the environmental impact of different air travel options based on the combination of the following aspects: aircraft type, engine type, seating configuration (Aircraft Label); airline environmental performance (Airline Label); number of legs of a trip, time, cost and environmental information (Flight Label). --- Methodology – The existing environmental label for aircraft considered resource depletion (fuel consumption), global warming (equivalent CO2 emission, including altitude-dependent NOx and aviation induced cloudiness), local air quality (NOx) and noise pollution. The data for determining fuel consumption and equivalent CO2 emissions was revised for existing aircraft and was extended with new aircraft types. Equivalent CO2 emissions were made dependent on the specific engine of the aircraft. The methodology for calculating CO2 equivalent emissions was refined with aviation induced cloudiness now being a function of fuel consumption. This improved aircraft label was used to evaluate the fleet of the 50 most important airlines with an airline label, which takes type and number of aircraft of an airline into consideration. Different methodologies of calculating the environmental impact of a flight used by flight booking engines were compared and discussed. Approaches for a multimodal trip score and a flight label were presented. --- Findings – An improved more accurate aircraft label was created. The database of aircraft, airline and engine combinations was extended. The environmental performance of over 50 airlines were calculated using the airline label, which resulted in an airline ranking. Different methods to incorporate a flight label into a flight booking engine were proposed based on the aircraft label approach. --- Research Limitations –The airline label does not consider airline specific data like the passenger/cargo load factor. Because of the nature of an environmental label to only focus on the most important criteria, there is no distinction made between the technical efficiency of different airlines. The local air pollution for turboprop aircraft could not be calculated due to a lack of publicly available data and missing access to the Swedish Defense Research Agency (FOI). --- Practical Implications – Passengers understand the most important criteria of a flight affecting its environmental burden. They can make an educated choice regarding the combination of aircraft, engine, airline and the chosen route. Obviously, a modern aircraft with an efficient engine, a ticket in the economy class and a direct flight should be chosen. --- Social Implications – The multimodal trip score does provide the user with the ability to choose a flight based on their personal preferences and circumstances. --- Originality – A logical trinity of the environmental labels in aviation plus an outlook to the multimodal trip score was not presented so far.

This is Part 1/2 of the ActiveHuman dataset! Part 2 can be found here. Dataset Description ActiveHuman was generated using Unity's Perception package. It consists of 175428 RGB images and their semantic segmentation counterparts taken at different environments, lighting conditions, camera distances and angles. In total, the dataset contains images for 8 environments, 33 humans, 4 lighting conditions, 7 camera distances (1m-4m) and 36 camera angles (0-360 at 10-degree intervals). The dataset does not include images at every single combination of available camera distances and angles, since for some values the camera would collide with another object or go outside the confines of an environment. As a result, some combinations of camera distances and angles do not exist in the dataset. Alongside each image, 2D Bounding Box, 3D Bounding Box and Keypoint ground truth annotations are also generated via the use of Labelers and are stored as a JSON-based dataset. These Labelers are scripts that are responsible for capturing ground truth annotations for each captured image or frame. Keypoint annotations follow the COCO format defined by the COCO keypoint annotation template offered in the perception package.

Folder configuration The dataset consists of 3 folders:

JSON Data: Contains all the generated JSON files. RGB Images: Contains the generated RGB images. Semantic Segmentation Images: Contains the generated semantic segmentation images.

Essential Terminology

Annotation: Recorded data describing a single capture. Capture: One completed rendering process of a Unity sensor which stored the rendered result to data files (e.g. PNG, JPG, etc.). Ego: Object or person on which a collection of sensors is attached to (e.g., if a drone has a camera attached to it, the drone would be the ego and the camera would be the sensor). Ego coordinate system: Coordinates with respect to the ego. Global coordinate system: Coordinates with respect to the global origin in Unity. Sensor: Device that captures the dataset (in this instance the sensor is a camera). Sensor coordinate system: Coordinates with respect to the sensor. Sequence: Time-ordered series of captures. This is very useful for video capture where the time-order relationship of two captures is vital. UIID: Universal Unique Identifier. It is a unique hexadecimal identifier that can represent an individual instance of a capture, ego, sensor, annotation, labeled object or keypoint, or keypoint template.

Dataset Data The dataset includes 4 types of JSON annotation files files:

annotation_definitions.json: Contains annotation definitions for all of the active Labelers of the simulation stored in an array. Each entry consists of a collection of key-value pairs which describe a particular type of annotation and contain information about that specific annotation describing how its data should be mapped back to labels or objects in the scene. Each entry contains the following key-value pairs:

id: Integer identifier of the annotation's definition. name: Annotation name (e.g., keypoints, bounding box, bounding box 3D, semantic segmentation). description: Description of the annotation's specifications. format: Format of the file containing the annotation specifications (e.g., json, PNG). spec: Format-specific specifications for the annotation values generated by each Labeler.

Most Labelers generate different annotation specifications in the spec key-value pair:

BoundingBox2DLabeler/BoundingBox3DLabeler:

label_id: Integer identifier of a label. label_name: String identifier of a label. KeypointLabeler:

template_id: Keypoint template UUID. template_name: Name of the keypoint template. key_points: Array containing all the joints defined by the keypoint template. This array includes the key-value pairs:

label: Joint label. index: Joint index. color: RGBA values of the keypoint. color_code: Hex color code of the keypoint skeleton: Array containing all the skeleton connections defined by the keypoint template. Each skeleton connection defines a connection between two different joints. This array includes the key-value pairs:

label1: Label of the first joint. label2: Label of the second joint. joint1: Index of the first joint. joint2: Index of the second joint. color: RGBA values of the connection. color_code: Hex color code of the connection. SemanticSegmentationLabeler:

label_name: String identifier of a label. pixel_value: RGBA values of the label. color_code: Hex color code of the label.

captures_xyz.json: Each of these files contains an array of ground truth annotations generated by each active Labeler for each capture separately, as well as extra metadata that describe the state of each active sensor that is present in the scene. Each array entry in the contains the following key-value pairs:

id: UUID of the capture. sequence_id: UUID of the sequence. step: Index of the capture within a sequence. timestamp: Timestamp (in ms) since the beginning of a sequence. sensor: Properties of the sensor. This entry contains a collection with the following key-value pairs:

sensor_id: Sensor UUID. ego_id: Ego UUID. modality: Modality of the sensor (e.g., camera, radar). translation: 3D vector that describes the sensor's position (in meters) with respect to the global coordinate system. rotation: Quaternion variable that describes the sensor's orientation with respect to the ego coordinate system. camera_intrinsic: matrix containing (if it exists) the camera's intrinsic calibration. projection: Projection type used by the camera (e.g., orthographic, perspective). ego: Attributes of the ego. This entry contains a collection with the following key-value pairs:

ego_id: Ego UUID. translation: 3D vector that describes the ego's position (in meters) with respect to the global coordinate system. rotation: Quaternion variable containing the ego's orientation. velocity: 3D vector containing the ego's velocity (in meters per second). acceleration: 3D vector containing the ego's acceleration (in ). format: Format of the file captured by the sensor (e.g., PNG, JPG). annotations: Key-value pair collections, one for each active Labeler. These key-value pairs are as follows:

id: Annotation UUID . annotation_definition: Integer identifier of the annotation's definition. filename: Name of the file generated by the Labeler. This entry is only present for Labelers that generate an image. values: List of key-value pairs containing annotation data for the current Labeler.

Each Labeler generates different annotation specifications in the values key-value pair:

BoundingBox2DLabeler:

label_id: Integer identifier of a label. label_name: String identifier of a label. instance_id: UUID of one instance of an object. Each object with the same label that is visible on the same capture has different instance_id values. x: Position of the 2D bounding box on the X axis. y: Position of the 2D bounding box position on the Y axis. width: Width of the 2D bounding box. height: Height of the 2D bounding box. BoundingBox3DLabeler:

label_id: Integer identifier of a label. label_name: String identifier of a label. instance_id: UUID of one instance of an object. Each object with the same label that is visible on the same capture has different instance_id values. translation: 3D vector containing the location of the center of the 3D bounding box with respect to the sensor coordinate system (in meters). size: 3D vector containing the size of the 3D bounding box (in meters) rotation: Quaternion variable containing the orientation of the 3D bounding box. velocity: 3D vector containing the velocity of the 3D bounding box (in meters per second). acceleration: 3D vector containing the acceleration of the 3D bounding box acceleration (in ). KeypointLabeler:

label_id: Integer identifier of a label. instance_id: UUID of one instance of a joint. Keypoints with the same joint label that are visible on the same capture have different instance_id values. template_id: UUID of the keypoint template. pose: Pose label for that particular capture. keypoints: Array containing the properties of each keypoint. Each keypoint that exists in the keypoint template file is one element of the array. Each entry's contents have as follows:

index: Index of the keypoint in the keypoint template file. x: Pixel coordinates of the keypoint on the X axis. y: Pixel coordinates of the keypoint on the Y axis. state: State of the keypoint.

The SemanticSegmentationLabeler does not contain a values list.

egos.json: Contains collections of key-value pairs for each ego. These include:

id: UUID of the ego. description: Description of the ego. sensors.json: Contains collections of key-value pairs for all sensors of the simulation. These include:

id: UUID of the sensor. ego_id: UUID of the ego on which the sensor is attached. modality: Modality of the sensor (e.g., camera, radar, sonar). description: Description of the sensor (e.g., camera, radar).

Image names The RGB and semantic segmentation images share the same image naming convention. However, the semantic segmentation images also contain the string Semantic_ at the beginning of their filenames. Each RGB image is named "e_h_l_d_r.jpg", where:

e denotes the id of the environment. h denotes the id of the person. l denotes the id of the lighting condition. d denotes the camera distance at which the image was captured. r denotes the camera angle at which the image was captured.

IntroductionFront-of-pack labels (FoPLs) are key public health tools that help consumers identify healthier food options. Although widely studied, little is known about their effectiveness in Saudi Arabia. This study aimed to determine the most understandable FoPL among five international systems to help Saudi consumers make healthier food choices.MethodsFrom January 1, 2022, to January 30, 2023, 2,509 Saudi consumers aged 18 years and above were recruited in public places across Riyadh. Participants were asked to select one product from sets of five food categories (bread, cheese, cereals, nuggets, and juice) with different nutritional profiles and then rank the products within each set based on their perceived nutritional quality. These tasks were first performed without any FoPL. Participants were then randomly assigned to one of the following five FoPL systems: Health Star Rating (HSR), Guideline Daily Amount (GDA), Multiple Traffic Lights (MTL), Chilean Warning Octagons (CWO), or Nutri-Score (NS), and asked to repeat the same tasks with the assigned label displayed on the packaging. Multivariate ordinal logistic regressions were performed to analyze whether changes in the scores of food choices and the ability to correctly rank the products were associated with the FoPL types, along with various socioeconomic and behavioral factors.ResultsThe analyses showed that participants improved their food choices depending on the FoPL format and the food category. Nutri-Score (NS) demonstrated a significant improvement in food choices across all food categories (OR = 1.96, 95% CI: 1.24 to 3.17, p = 0.003), particularly for nuggets (OR = 2.18, 95% CI: 1.16 to 3.17, p = 0.038) and cereals (OR = 2.16, 95% CI: 1.28 to 4.53, p = 0.001), compared to the GDA label. All FoPL types resulted in a greater proportion of correct responses in the ranking task compared to the no-label condition. Furthermore, NS emerged as the most influential FoPL in enhancing participants’ understanding of nutritional quality, significantly improving their ability to correctly rank products across all food categories (OR = 5.81, 95% CI: 2.92 to 7.28, p

Datasets accompanying the study “A Taxonomy of Tools and Approaches for FAIRification” on the tools and approaches emerging from stakeholders’ experiences adopting the FAIR principles in practice. Datasets: queryResults.csv Description The dataset consists of the query results returned by OpenAIRE Explore and defines the corpus at the base of our study. Structure 11 columns: Query Type of query entered FAIR, FAIRification (all fields) OpenAIRE subjects (subject) Result Type [OpenAIRE label] Type of the research output (publication|data|software|other) Title [OpenAIRE label] Authors [OpenAIRE label] Publication Year [OpenAIRE label] DOI [OpenAIRE label] Download from [OpenAIRE label] Type [OpenAIRE label] Subtype of the research output Journal [OpenAIRE label] Funder|Project Name (GA Number) [OpenAIRE label] Access [OpenAIRE label] Access rights publicationsTools.csv Description The dataset pairs the tools/services extracted from the corpus to their respective source. Structure 2 columns: source reference to the publication or software citation name name of the tool/service/technology toolsAll.csv Description The dataset lists all the unique tool/service entries, distinguishing between those that were considered relevant for the study (further categorised into tools, technologies or services) and those that were excluded. Structure 3 columns: entryType entry categorisation (tool|service|technology|excluded) name name of the tool/service/technology URL URL of the tool/service web page or description toolsType.csv Description Classification of the tools/services/technologies into the study-defined classes. Structure 19 columns: name name of the tool/service/technology URL URL of the tool/service web page or description GUPRI helper - GUPRI creation and management service ‘class - subclass’ of the tool/service/technology GUPRI helper - GUPRI Indexing and discovery service ‘class - subclass’ of the tool/service/technology Metadata helper - Metadata editor ‘class - subclass’ of the tool/service/technology Metadata helper - Metadata extractor ‘class - subclass’ of the tool/service/technology Metadata helper - Metadata tracker ‘class - subclass’ of the tool/service/technology Metadata helper - Metadata validator ‘class - subclass’ of the tool/service/technology Metadata helper - Metadata assistant ‘class - subclass’ of the tool/service/technology Indexing and discovery service - registry ‘class - subclass’ of the tool/service/technology Indexing and discovery service - repository ‘class - subclass’ of the tool/service/technology Indexing and discovery service - Indexing and discovery service finder ‘class - subclass’ of the tool/service/technology Converter - metadata ‘class - subclass’ of the tool/service/technology Converter - data ‘class - subclass’ of the tool/service/technology Licence helper ‘class’ of the tool/service/technology Assessment tool - automated ‘class - subclass’ of the tool/service/technology Assessment tool - manual ‘class - subclass’ of the tool/service/technology Assessment tool - Assessment tool finder ‘class - subclass’ of the tool/service/technology DMP tool ‘class’ of the tool/service/technology toolsFAIR.csv Description The dataset relates the tool/service/technology to the FAIR principles it enables. Structure 12 columns: name name of the tool/service/technology URL URL of the tool/service web page or description F1 reference to the FAIR principle F2 reference to the FAIR principle F3 reference to the FAIR principle F4 reference to the FAIR principle A generic reference to the accessibility principles (see the paper) I1 reference to the FAIR principle I3 reference to the FAIR principle R1.1 reference to the FAIR principle R1.2 reference to the FAIR principle R1.3 reference to the FAIR principle toolsScope.csv Description Since the FAIR principles have been specified for different types of resources ((meta)data, semantic artefacts, software and workflows), the dataset correlates the tool/service/technology and the types of FAIR-specific resources it covers. Structure 6 columns: name name of the tool/service/technology URL URL of the tool/service web page or description (meta)data reference to the FAIR-specific resource semantic artefact reference to the FAIR-specific resource software reference to the FAIR-specific resource workflow reference to the FAIR-specific resource toolsDomain.csv Description Classification of the tools/services/technologies into the Frascati framework-defined domains. Structure 9 columns: name name of the tool/service/technology URL URL of the tool/service web page or description cross-domain domain Agricultural and veterinary sciences domain Engineering and technology domain Humanities and the arts domain Medical and health sciences domain Natural sciences domain Social sciences domain

Variable Data Booklet Label for Pharma Market Outlook

According to our latest research, the global Variable Data Booklet Label for Pharma market size reached USD 1.42 billion in 2024, with a robust CAGR of 6.1% expected throughout the forecast period. By 2033, the market is projected to attain a value of USD 2.41 billion, driven by the increasing demand for advanced labeling solutions in the pharmaceutical sector. The market’s growth is primarily fueled by stringent regulatory requirements, the surge in pharmaceutical production, and the need for comprehensive, multi-lingual, and patient-centric labeling formats.

One of the most significant growth factors influencing the Variable Data Booklet Label for Pharma market is the ever-tightening regulatory landscape across global pharmaceutical markets. Regulatory bodies such as the FDA, EMA, and other regional authorities mandate the inclusion of extensive product information, patient safety instructions, multilingual content, and serialization data on pharmaceutical packaging. Booklet labels, especially those with variable data capability, allow manufacturers to efficiently incorporate critical, product-specific information while maintaining compliance with evolving guidelines. Furthermore, the rise in global drug recalls and counterfeiting incidents has compelled pharmaceutical companies to seek labeling solutions that support track-and-trace features, serialization, and tamper-evidence, all of which are efficiently addressed by advanced variable data booklet labels.

Another major driver is the escalating complexity of pharmaceutical products and their distribution networks. With the expansion of personalized medicine, biologics, and clinical trials, there is a growing need to deliver highly detailed product and usage information tailored to specific batches, regions, or patient groups. Variable data booklet labels enable pharmaceutical manufacturers to provide customized content, such as dosage instructions, adverse effect warnings, and multi-country regulatory compliance, all within a compact labeling format. Moreover, the increasing adoption of digital printing technologies has significantly enhanced the flexibility, speed, and cost-effectiveness of producing variable data booklet labels, allowing for agile response to market and regulatory changes.

The global push towards patient-centric healthcare and improved medication adherence further contributes to the growth of the Variable Data Booklet Label for Pharma market. Booklet labels offer ample space for patient education, medication guides, and detailed instructions, thus reducing medication errors and improving health outcomes. Pharmaceutical companies are increasingly leveraging booklet labels to communicate vital information in multiple languages, address diverse patient demographics, and comply with region-specific regulations. Additionally, sustainability initiatives are prompting manufacturers to adopt eco-friendly materials and printing processes, enhancing the appeal of booklet labels that combine compliance, patient safety, and environmental responsibility.

From a regional perspective, North America and Europe continue to dominate the Variable Data Booklet Label for Pharma market, thanks to their advanced pharmaceutical industries, high regulatory standards, and early adoption of innovative labeling technologies. However, the Asia Pacific region is witnessing the fastest growth, driven by the rapid expansion of pharmaceutical manufacturing, increasing regulatory scrutiny, and rising exports. Latin America and the Middle East & Africa are also emerging as significant markets, buoyed by improving healthcare infrastructure, growing pharmaceutical consumption, and the need for compliance with international labeling standards. The interplay of these regional dynamics is shaping the global competitive landscape, with multinational and local players investing in advanced printing technologies and sustainable materials to capture market share.

Product Type Analysis

The Product Type segment of the Variable Data Booklet Label for Pharma market encompasses multi-page booklet labels, fold-out booklet labels, peel-and-read booklet labels, and other specialized formats. Multi-page booklet labels remain the most widely adopted product type, particularly for prescription medications and products requiring extensive regulatory and usage information. These labels can accommodate multiple languages, detailed