Video Object Tracking

## Overview

Video Object Tracking is a dataset for object detection tasks - it contains Boundary annotations for 1,672 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [CC BY 4.0 license](https://creativecommons.org/licenses/CC BY 4.0).

Monitoring animals in their natural habitat is essential for the advancement of animal behavioural studies, especially in pollination studies. We present a novel hybrid detection and tracking algorithm "HyDaT" to monitor unmarked insects outdoors. Our software can detect an insect, identify when a tracked insect becomes occluded from view and when it re-emerges, determine when an insect exits the camera field of view, and our software assembles a series of insect locations into a coherent trajectory. The insect detecting component of the software uses background subtraction and deep learning-based detection together to accurately and efficiently locate the insect.This dataset includes videos of honeybees foraging in two ground-covers Scaevola and Lamb's-ear, comprising of complex background detail, wind-blown foliage, and honeybees moving into and out of occlusion beneath leaves and among three-dimensional plant structures. Honeybee tracks and associated outputs of experiments extracted using HyDaT algorithm are included in the dataset. The dataset also contains annotated images and pre-trained YOLOv2 object detection models of honeybees.

The dataset contains frames extracted from videos with dogs on the streets. Each frame is accompanied by bounding box that specifically tracks the dog in the image. The dataset provides a valuable resource for advancing computer vision tasks, enabling the development of more accurate and effective solutions for monitoring and understanding dog behavior in urban settings.

Football Video Tracking 2

## Overview

Football Video Tracking 2 is a dataset for object detection tasks - it contains Football Players annotations for 9,026 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [CC BY 4.0 license](https://creativecommons.org/licenses/CC BY 4.0).

The collection of overhead video frames, capturing various types of vehicles traversing a roadway. The dataset inculdes light vehicles (cars) and heavy vehicles (minivan).

This dataset contains 11,352 overhead surveillance videos recorded in diverse indoor and outdoor scenarios across various time periods. Captured from top-down views, the data is suitable for tasks such as human detection, pedestrian tracking, people counting, and human body attribute analysis. It is ideal for developing and benchmarking computer vision models in surveillance and security applications. All data are legally sourced and fully compliant with GDPR, CCPA, and PIPL.

This dataset was recorded for evaluating the accuracies of a GPS- and a camera-based object tracking. For this purpose, one person was equipped with several (equal) GPS loggers. Further, a camera was used to observe this person following a predefined course. Thus, this dataset contains the GPS trajectories provided by the loggers, the corresponding video and some meta information describing the setup (time synchronization, field coordinates, waypoint coordinates, camera location and the homography matrix).

The dataset comprises of annotated video frames from positioned in a public space camera. The tracking of each individual in the camera's view has been achieved using the rectangle tool in the Computer Vision Annotation Tool (CVAT).

Cargo Detection Tracking Video

## Overview

Cargo Detection Tracking Video is a dataset for object detection tasks - it contains Cargo YRlt annotations for 420 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [CC BY 4.0 license](https://creativecommons.org/licenses/CC BY 4.0).

Medical Staff People Tracking - Object Detection dataset

The dataset contains a collection of frames extracted from videos captured within a hospital environment. The bounding boxes are drawn around the doctors, nurses, and other people who appear in the video footage.

💴 For Commercial Usage: To discuss your requirements, learn about the price and buy the dataset, leave a request on our website to buy the dataset

The dataset can be used for computer vision in healthcare settings and the development of systems that monitor medical staff activities, patient flow, analyze wait times, and assess the efficiency of hospital processes.

https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F12421376%2F923816c1040f503bbe0cb20c8d1f2446%2Fezgif.com-optimize.gif?generation=1695376624951041&alt=media" alt="">

Dataset structure

The dataset consists of 2 folders with frames from the video from a hospital. Each folder includes: - images: folder with original frames from the video, - boxes: visualized data labeling for the images in the previous folder, - .csv file: file with id and path of each frame in the "images" folder, - annotations.xml: contains coordinates of the bounding boxes, created for the original frames

🧩 This is just an example of the data. Leave a request here to learn more

Data Format

Each frame from images folder is accompanied by an XML-annotation in the annotations.xml file indicating the coordinates of the bounding boxes for people tracking. For each point, the x and y coordinates are provided.

Classes:

• doctor - doctor in the frame
• nurse - nurse in the frame
• others - other people (not medical staff)

Example of the XML-file

https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F12421376%2F778b3f19625b76cdc5454d58258fa0aa%2Fcarbon%20(1).png?generation=1695995011699193&alt=media" alt="">

Object tracking might be made in accordance with your requirements.

🚀 You can learn more about our high-quality unique datasets here

keywords: human detection dataset, object detection dataset, people tracking dataset, tracking human object interactions, human Identification tracking dataset, people detection annotations, human trafficking dataset, deep learning object tracking, multi-object tracking dataset, labeled web tracking dataset, large-scale object tracking dataset, image dataset, classification, object detection, medical data, doctors, nurses

The global Automatic Video Tracker market is experiencing robust growth, driven by increasing demand for advanced surveillance and security systems across various sectors. The market size in 2025 is estimated at $2.5 billion, exhibiting a Compound Annual Growth Rate (CAGR) of 15% from 2025 to 2033. This significant growth is fueled by several key factors, including the rising adoption of AI-powered video analytics, the need for enhanced situational awareness in critical infrastructure like airports and transportation hubs, and the increasing prevalence of smart cities initiatives. The market is segmented by application (security, industrial automation, defense), technology (optical, infrared), and region (North America, Europe, Asia-Pacific). Key players such as JK Paper Ltd.(Delopt), AMETEK, Inc.(Abaco Systems), and others are actively innovating and expanding their product portfolios to capitalize on the expanding market opportunities. Further market expansion is expected to be driven by technological advancements in image processing, sensor technology, and the integration of cloud computing for data storage and analysis. The rising adoption of low-cost high-performance sensors and the development of more sophisticated algorithms for object detection and tracking are also contributing to market growth. However, challenges such as high initial investment costs and the need for specialized expertise to implement and maintain these systems could act as potential restraints. The forecast period of 2025-2033 presents significant opportunities for market players to leverage emerging technologies and cater to the increasing demand for reliable and efficient automatic video tracking solutions across diverse applications.

1. The DVS recordings for event-stream pattern tracking:"DAVIS recordings and groundtruth.zip" and "DVS recordings and groundtruth.zip"2. results displayed by videos

This dataset is used to adapt DeepLabCut for Human motion tracking.

  Structure of the dataset

videos contains 100+ videos of 4 candidates recorded during a game of darts. labeled-data contains labels on the corresponding frames of the videos. These labels are used to adapt DeepLabCut for human motion tracking. Under labeled-data there are 2 folders for every video. video_name has all the relevant frames extracted from the video, xy coordinates of the labels in the csv file and the… See the full description on the dataset page: https://huggingface.co/datasets/GT-Neuronext/human-motion-tracking-deeplabcut.

A Car Detection and Tracking Dataset is a curated collection of images or videos, often with accompanying metadata, designed to train and evaluate machine learning models that detect and track vehicles within visual data. These datasets are critical for developing algorithms used in autonomous driving, traffic monitoring, and intelligent transportation systems.

Key Features:

1. Data Types:

   • Images: Static images capturing various scenes such as highways, urban streets, parking lots, etc.
   • Videos: Sequences of frames allowing temporal analysis for tracking cars over time.

2. Annotations:

   • Bounding Boxes: Rectangular boxes around detected vehicles, specifying the location and size of each car in the image or video.
   • Class Labels: Indication of whether an object is a car, truck, or other vehicle types, sometimes including more detailed classifications (e.g., sedans, SUVs).
   • Tracking IDs: Unique identifiers for each vehicle across frames in a video, enabling the tracking of a specific car as it moves.
   • Environmental Metadata: Information such as time of day, weather conditions, and camera angle, which can affect detection and tracking performance.

3. Diversity:

   • Variety of Scenes: Data from different geographical locations, times of day, and weather conditions to ensure robust model training.
   • Vehicle Types: Inclusion of different vehicle sizes, colors, and models to reflect real-world diversity.
   • Occlusions and Overlaps: Scenarios where vehicles are partially hidden or overlapping, adding complexity to the detection and tracking tasks.

4. Application Scenarios:

   • Autonomous Driving: Essential for training self-driving cars to recognize and respond to other vehicles on the road.
   • Traffic Surveillance: Used for monitoring and analyzing traffic patterns, detecting violations, and improving traffic management.
   • Smart Cities: Facilitates the development of smart infrastructure that interacts with vehicles, such as adaptive traffic lights and congestion detection systems.

5. Challenges:

   • Varied Lighting and Weather Conditions: Data often includes challenging scenarios like low-light, rain, snow, and shadows, which can affect detection accuracy.
   • High Density Traffic**: Scenes with numerous vehicles in close proximity, increasing the difficulty of accurate detection and tracking.
   • Camera Variability**: Data from different types of cameras (e.g., dashcams, surveillance cameras) with varying resolutions and frame rates.

Drone Video AI Object Tracking for Search Market Outlook

According to our latest research, the global Drone Video AI Object Tracking for Search market size reached USD 1.18 billion in 2024, driven by increasing demand for real-time situational awareness and operational efficiency across diverse sectors. The market is forecasted to grow at a robust CAGR of 17.6% from 2025 to 2033, reaching an estimated USD 5.02 billion by 2033. This impressive growth trajectory is primarily attributed to advancements in AI-powered analytics, the proliferation of drone adoption, and the expanding use cases in critical applications such as search and rescue, disaster management, and public safety.

A major growth factor for the Drone Video AI Object Tracking for Search market is the rapid evolution of artificial intelligence and machine learning algorithms, which have significantly enhanced the accuracy and speed of object detection and tracking in complex environments. The integration of deep learning and computer vision technologies into drone systems enables the automated identification and real-time tracking of people, vehicles, and objects, even in challenging conditions such as low visibility or rugged terrain. This technological leap has revolutionized search operations by reducing human error, minimizing response times, and improving the overall effectiveness of search missions. Additionally, the increasing availability of high-resolution cameras and advanced sensors on drones has further bolstered the capabilities of AI-based object tracking systems, allowing for more precise and reliable outcomes in mission-critical scenarios.

Another significant driver is the expanding application of drone-based AI object tracking in public safety and emergency response operations. Search and rescue teams, law enforcement agencies, and disaster management authorities are increasingly leveraging drones equipped with AI-powered video analytics to conduct rapid area sweeps, locate missing persons, and monitor evolving situations in real time. The ability to deploy drones quickly in inaccessible or hazardous areas, combined with the power of AI-driven object tracking, has proven invaluable in saving lives and mitigating damage during natural disasters, accidents, and security incidents. Moreover, the cost-effectiveness and scalability of drone solutions compared to traditional manned operations have encouraged widespread adoption, particularly among resource-constrained agencies and organizations.

The market is also experiencing growth due to the rising demand for environmental monitoring and wildlife conservation applications. Environmental agencies and non-governmental organizations are utilizing drones with AI object tracking capabilities to monitor animal populations, track illegal poaching activities, and assess ecological changes in remote or protected areas. The ability to collect and analyze large volumes of video data autonomously enhances the efficiency and accuracy of research and conservation efforts. Furthermore, the commercial sector, including infrastructure inspection, logistics, and agriculture, is increasingly adopting drone-based AI object tracking for asset monitoring, perimeter security, and operational optimization, thereby expanding the addressable market and fueling further innovation.

From a regional perspective, North America currently dominates the Drone Video AI Object Tracking for Search market, accounting for the largest market share in 2024, followed closely by Europe and Asia Pacific. The presence of leading technology providers, robust regulatory frameworks, and significant investments in public safety and defense applications have positioned North America as a key growth engine. However, the Asia Pacific region is expected to witness the highest CAGR during the forecast period, driven by rapid technological adoption, government initiatives for disaster management, and increasing demand from emerging economies such as China and India. Europe continues to demonstrate strong growth potential, supported by active R&D efforts and cross-border collaborations in security and environmental monitoring.

Component Analysis

The Component segment of the Drone Video AI Object Tracking for Search market is categorized into Software, Hardware, and Services, each playing a critical role in the ecosystem. Software solutions, comprising AI algorithms, computer vision modules, and data analytics platforms, are the back

We collected and annotated a dataset containing 105,544 annotated vehicle instances from 24700 image frames within seven different videos, sourced online under creative commons license. The video frames are annotated using DarkLabel tool. In the interest of reusability and generalisation of the deep learning model, we consider the diversity within the collected dataset. This diversity includes changes of lighting amongst the video, as well as other factors such as weather conditions, angle of observation, varying speed of the moving vehicles, traffic flow, and road conditions etc. The videos collected obviously include stationary vehicles, to perform the validation of stopped vehicle detection method. It can be noticed that the road conditions (e.g., motorways, city, country roads), directions, data capture timings and camera views, vary in the dataset producing annotated dataset with diversity. the dataset may have several uses such as vehicle detection, vehicle identification, stopped vehicle detection on smart motorways and local roads (smart city applications) and many more.

Automatic Video Tracker Market Outlook

The global automatic video tracker market size is projected to grow from USD 1.2 billion in 2023 to USD 2.5 billion by 2032, reflecting a CAGR of 8.5% during the forecast period. The substantial growth in this market can be attributed to the increasing demand for enhanced surveillance systems, improvements in video analytics technology, and a growing emphasis on public safety and security across various sectors.

One of the primary growth factors driving the automatic video tracker market is the rising need for advanced surveillance systems. With security concerns escalating globally, both government and private sectors are investing heavily in state-of-the-art video tracking technology to ensure public safety. These systems offer real-time monitoring, automatic tracking of moving objects, and integration with other security measures, making them indispensable for modern surveillance infrastructures.

Another significant factor contributing to market growth is technological advancements in video analytics. Innovations such as AI and machine learning have significantly enhanced the capabilities of automatic video trackers, making them more accurate and efficient. These advancements enable video trackers to identify and analyze objects with higher precision, reducing false alarms and improving overall security measures. As technology continues to evolve, the demand for sophisticated video tracking solutions is expected to surge.

The increasing adoption of automatic video trackers in various sectors such as sports, entertainment, and healthcare further propels market growth. In sports and entertainment, these trackers are used for performance analysis and audience engagement, while in healthcare, they aid in patient monitoring and operational efficiency. The versatility of automatic video trackers in diverse applications underscores their market potential and drives their adoption across multiple industries.

From a regional perspective, North America currently dominates the automatic video tracker market, followed by Europe and Asia Pacific. The strong presence of key market players, coupled with significant investments in security infrastructure, positions North America as a leading region. However, the Asia Pacific region is expected to witness the highest growth rate during the forecast period, driven by rapid urbanization, increasing public safety concerns, and government initiatives to enhance surveillance systems.

Component Analysis

The automatic video tracker market is segmented by components, including software, hardware, and services. The software segment holds a significant market share due to the critical role of advanced software solutions in video tracking systems. These software solutions encompass a range of functionalities such as motion detection, object recognition, and real-time analytics, which are essential for effective video tracking. With continuous software innovations and updates, this segment is poised for robust growth.

The hardware segment is also crucial to the market, encompassing cameras, sensors, and processing units. High-definition cameras and advanced sensors have become integral to video tracking systems, providing the necessary data for accurate tracking and analysis. The growing demand for high-quality, durable hardware components ensures that this segment remains a vital part of the market. Additionally, improvements in hardware technology, such as the development of low-power consumption devices and enhanced image processing capabilities, contribute to the segment's growth.

The services segment includes maintenance, installation, and consulting services related to automatic video trackers. As the adoption of these systems increases, the demand for specialized services to support their implementation and operation also rises. Service providers play a crucial role in ensuring the seamless functioning of video tracking systems, offering expertise in system integration, troubleshooting, and optimization. This segment is expected to grow steadily as organizations seek to maximize the efficiency and reliability of their video tracking solutions.

Overall, the component analysis reveals that each segment—software, hardware, and services—plays an integral role in the automatic video tracker market. These components work synergistically to deliver comprehensive video tracking solutions that cater to the diverse needs of various industries. As technology continues to advance, the interplay between these compon

This project contains annotated and unannotated behavioural video footage of meerkats at Wellington Zoo. Data is annotated as bounding boxes with a behaviour class for each frame.

Abstract: Recording animal behaviour is an important step in evaluating the well-being of animals and further understanding the natural world. Current methods for documenting animal behaviour within a zoo setting, such as scan sampling, require excessive human effort, are unfit for around-the-clock monitoring, and may produce human-biased results. Several animal datasets already exist that focus predominantly on wildlife interactions, with some extending to action or behaviour recognition. However, there is limited data in a zoo setting or data focusing on the group behaviours of social animals. We introduce a large meerkat (Suricata Suricatta) behaviour recognition video dataset with diverse annotated behaviours, including group social interactions, tracking of individuals within the camera view, skewed class distribution, and varying illumination conditions. This dataset includes videos from two positions within the meerkat enclosure at the Wellington Zoo (Wellington, New Zealand), with 848,400 annotated frames across 20 videos and 15 unannotated videos.

Associated annotations: https://meerkat-dataset.github.io/Annotations.zip More information: https://meerkat-dataset.github.io/

Ground Truth Para Tracking YOLOV8 Video 633

## Overview

Ground Truth Para Tracking YOLOV8 Video 633 is a dataset for object detection tasks - it contains Labeled Cars Trucks Buses annotations for 649 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [Public Domain license](https://creativecommons.org/licenses/Public Domain).

This Project consists of two datasets, both of aerial images and videos of dolphins, being taken by drones. The data was captured from few places (Italy and Israel coast lines).

The aim of the project is to examine automated dolphins detection and tracking from aerial surveys.

The project description, details and results are presented in the paper (link to the paper).

Each dataset was organized and set for a different phase of the project. Each dataset is located in a different zip file:

1. Detection - Detection.zip

2. Tracking - Tracking.zip

Further information about the datasets' content and annotation format is below.

• In aim to watch each file content, use the preview option, in addition a description appears later on this section.

Detection Dataset

This dataset contains 1125 aerial images, while an image can contain several dolphins.

The detection phase of the project is done using RetinaNet, supervised deep learning based algorithm, with the implementation of Keras RetinaNet. Therefore, the data was divided into three parts - Train, Validation and Test. The relations is 70%, 15%, 15% respectively.

The annotation format follows the requested format of that implementation (Keras RetinaNet). Each object, which is a dolphin, is annotated as a bounding box coordinates and a class. For this project, the dolphins were not distinguished into species, therefore, a dolphin object is annotated as a bounding box, and classified as a 'Dolphin'. Detection zip file includes:

A folder for each - Train, Validation and Test subsets, which includes the images

An annotations CSV file for each subset

A class mapping csv file (one for all the subsets).

*The annotation format is detailed in Annotation section.

Detection zip file content:

Detection |——————train_set (images) |——————train_set.csv |——————validation_set (images) |——————train_set.csv |——————test_set (images) |——————train_set.csv └——————class_mapping.csv

Tracking

This dataset contains 5 short videos (10-30 seconds), which were trimmed from a longer aerial videos, captured from a drone.

The tracking phase of the project is done using two metrics:

VIAME application, using the tracking feature

Re3: Real-Time Recurrent Regression Networks for Visual Tracking of Generic Objects, by Daniel Gordon. For this project, the author's Tensorflow implementation is being used

Both metrics demand the videos' frames sequence as an input. Therefore, the videos' frames were extracted. The first frame was annotated manually for initialization, and the algorithms track accordingly. Same as the Detection dataset, each frame can includes several objects (dolphins).

For annotation consistency, the videos' frames sequences were annotated similar to the Detection Dataset above, (details can be found in Annotation section). Each video's frames annotations separately. Therefore, Tracking zip file contains a folder for each video (5 folders in total), named after the video's file name.

Each video folder contains:

Frames sequence directory, which includes the extracted frames of the video

An annotations CSV file

A class mapping CSV file

The original video in MP4 format

The examined videos description and details are displayed in 'Videos Description.xlsx' file. Use the preview option for displaying its content.

Tracking zip file content:

Tracking |——————DJI_0195_trim_0015_0045 | └——————frames (images) | └——————annotations_DJI_0195_trim_0015_0045.csv | └——————class_mapping_DJI_0195_trim_0015_0045.csv | └——————DJI_0195_trim_0015_0045.MP4 |——————DJI_0395_trim_0010_0025 | └——————frames (images) | └——————annotations_DJI_0395_trim_0010_0025.csv | └——————class_mapping_DJI_0395_trim_0010_0025.csv | └——————DJI_0195_trim_0015_0045.MP4 |——————DJI_0395_trim_00140_00150 | └——————frames (images) | └——————annotations_DJI_0395_trim_00140_00150.csv | └——————class_mapping_DJI_0395_trim_00140_00150.csv | └——————DJI_0395_trim_00140_00150.MP4 |——————DJI_0395_trim_0055_0085 | └——————frames (images) | └——————annotations_DJI_0395_trim_0055_0085.csv | └——————class_mapping_DJI_0395_trim_0055_0085.csv | └——————DJI_0395_trim_0055_0085.MP4 └——————HighToLow_trim_0045_0070 └—————frames (images) └—————annotations_HighToLow_trim_0045_0070.csv └—————class_mapping_HighToLow_trim_0045_0070.csv └—————HighToLow_trim_0045_0070.MP4

Annotations format

Both datasets have similar annotation format which is described below. The data annotation format, of both datasets, follows the requested format of Keras RetinaNet Implementation, which was used for training in the Dolphins Detection phase of the project.

Each object (dolphin) is annotated by a bounding box left-top and right-bottom coordinates and a class. Each image or frame can includes several objects. All data was annotated using Labelbox application.

For each subset (Train, Validation and Test of Detection dataset, and each video of Tracking Dataset) there are two corresponded CSV files:

Annotations CSV file

Class mapping CSV file

Each line in the Annotations CSV file contains an annotation (bounding box) in an image or frame. The format of each line of the CSV annotation is:

path/to/image.jpg - a path to the image/frame

x1, y1 - image coordinates of the left upper corner of the bounding box

x2, y2 - image coordinates of the right bottom corner of the bounding box

class_name - class name of the annotated object

path/to/image.jpg,x1,y1,x2,y2,class_name

An example from train_set.csv:

.\train_set\1146_20170730101_ce1_sc_GOPR3047 103.jpg,506,644,599,681,Dolphin .\train_set\1146_20170730101_ce1_sc_GOPR3047 103.jpg,394,754,466,826,Dolphin .\train_set\1147_20170730101_ce1_sc_GOPR3047 104.jpg,613,699,682,781,Dolphin .\train_set\1147_20170730101_ce1_sc_GOPR3047 104.jpg,528,354,586,443,Dolphin .\train_set\1147_20170730101_ce1_sc_GOPR3047 104.jpg,633,250,723,307,Dolphin

This defines a dataset with 2 images:

1146_20170730101_ce1_sc_GOPR3047 103.jpg which contains 2 objects classified as 'Dolphin'

1146_20170730101_ce1_sc_GOPR3047 104.jpg which contains 3 objects classified as 'Dolphin'

Each line in the Class Mapping CSV file contains a mapping:

class_name,id

An example:

Dolphin,0