Movement

## Overview

Movement is a dataset for object detection tasks - it contains Sleeping annotations for 1,058 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).

Movement

## Overview

Movement is a dataset for object detection tasks - it contains Car Truck Bus MotorCycle annotations for 1,374 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 [MIT license](https://creativecommons.org/licenses/MIT).

The dataset contains tri-axial accelerometer and tri-axial gyroscope readings from the four IMUs and labels. There are three sub-datasets, which have different ground-truth labelling configurations, included in this dataset. Please note that the labelling is subjective to the mother's perception. The dataset, as a whole, contains recordings spanning 14 weeks from 26th week to 39th week and in total, about 71 hours of recordings.

The three sub-datasets included are:

Sub-dataset One: In this sub-dataset, only the occurrence of the particular type of fetal movement known as the fetal kick is considered for ground-truth labelling.

Sub-dataset Two: All types of fetal movement felt by the mother -- including trunk movement, isolated limb movement, and general body movement -- were considered for ground truth-labelling as fetal movements in this sub-dataset.

Sub-dataset Three: In this sub-dataset, the emphasis was given to the classification of different types of fetal movements. Three types of fetal movements are labelled: trunk movement, isolated limb movement, and general body movement.

Additional data are provided in three additional 'csv' files, which contains the record number, the Period of Amenorrhoea (POA), start time, and end time of each recording. Also, additional details about the mother and the baby are provided in the README file.

For more details, refer to the README.pdf.

This project is to determine horizontal and vertical movement patterns of two jellyfish species in Hood Canal, in relation to environmental variables. It is being conducted by NMFS scientists in collaboration with a NOAA Hollings Scholar; we also are making use of publicly available oceanographic data from the University of Washington. We used acoustic tags and receivers to track jellyfish move...

The electrooculography signal is widely used to analyze eye movements

Dataset

This dataset was created by oummuo

Contents

It contains the following files:

Comprehensive collection of movement specifications and data

Comprehensive financial and analytical metrics for Movement, including key performance indicators, market data, and ecosystem analytics.

Container Movement

## Overview

Container Movement is a dataset for object detection tasks - it contains Container Number annotations for 480 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).

The LTMM database contains 3-day 3D accelerometer recordings of 71 elder community residents, used to study gait, stability, and fall risk.

The field of movement ecology has rapidly grown during the last decade, with important advancements in tracking devices and analytical tools that have provided unprecedented insights into where, when, and why species move across a landscape. Although there has been an increasing emphasis on making animal movement data publicly available, there has also been a conspicuous dearth in the availability of such data on large carnivores. Globally, large predators are of conservation concern. However, due to their secretive behavior and low densities, obtaining movement data on apex predators is expensive and logistically challenging. Consequently, the relatively small sample sizes typical of large carnivore movement studies may limit insights into the ecology and behavior of these elusive predators. The aim of this initiative is to make available to the conservation-scientific community a dataset of 134,690 locations of jaguars (Panthera onca) collected from 117 individuals (54 males and 63 fe...

This dataset contains key characteristics about the data described in the Data Descriptor An Asian-centric human movement database capturing activities of daily living. Contents:

    1. human readable metadata summary table in CSV format


    2. machine readable metadata file in JSON format

Records represent the organization details for the population of transnationally organized activist organizations.

A database was created in .XLSX and .CSV formats containing the processing of an EMG signal and the position and angle error during the execution of three dynamic tasks based on the three-dimensional movement of the upper limb. This data was recorded from the quantification of the hand position error.

Summary

The weDraw Movement Corpus consists of three datasets: the weDraw-1 Movement Dataset (w1MD), the VI-weDraw Movement Dataset (VIwMD), and the weDraw-Games Movement Dataset (wGMD). The w1MD and VIwMD datasets were collected from children (with visual impairment, for the VIwMD dataset) based on game-like (i.e. fun) activities inspired by qualitative pedagogical studies. The wGMD dataset was based on weDraw serious games (Cartesian Garden, Angles Shapes and Fraction Musical Games) in both schools for children with visual impairment and mainstream schools.

The three datasets were built using video cameras, a Microsoft Kinect sensor, and the Notch wearable sensors. However, only the motion capture data from these sensors (and not raw videos) are open.

Please see the corpus README for more details.

How To Cite and Where to Get More Details

Please cite the following papers in all publications that result from use of the weDraw-1 Movement Dataset:

Olugbade, T. A., Newbold, J., Johnson, R., Volta, E., Alborno, P., Dillon, M., Volpe, G., Bianchi-Berthouze, N. Automatic Detection of Reflective Thinking in Mathematical Problem Solving based on Unconstrained Bodily Exploration, https://arxiv.org/abs/1812.07941.

More details about the weDraw-1 Movement Dataset can also be found there.

This dataset was originally created by Justin Henke and Reginald Viray. To see the current project, which may have been updated since this version, please go here: https://universe.roboflow.com/psi-dhxqe/psi-rossville-pano.

This dataset is part of RF100, an Intel-sponsored initiative to create a new object detection benchmark for model generalizability.

Access the RF100 Github repo: https://github.com/roboflow-ai/roboflow-100-benchmark

Cattle Movement Detection

## Overview

Cattle Movement Detection is a dataset for instance segmentation tasks - it contains Cattle Body Head annotations for 222 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 project’s objective is to document movement patterns and survival rates of Chinook salmon, steelhead, green sturgeon, and other fish from several sources in the Central Valley of California. Juvenile salmonids from hatcheries or wild caught are implanted with small acoustic transmitters and the location of the fish are recorded on receivers that are placed throughout the watershed from Redding to the Golden Gate. Over 70 receiver locations with over 150 receivers monitor the movement of these fish. These receivers record the date, time, and unique identification number of transmitters that pass within listening range of the receivers. The first acoustic tagging studies began in 2006 and continue today.

Dataset Card for truck-movement

** The original COCO dataset is stored at dataset.tar.gz**

  Dataset Summary

truck-movement

  Supported Tasks and Leaderboards

object-detection: The dataset can be used to train a model for Object Detection.

  Languages

English

  Dataset Structure





  Data Instances

A data point comprises an image and its object annotations. { 'image_id': 15, 'image': <PIL.JpegImagePlugin.JpegImageFile image mode=RGB… See the full description on the dataset page: https://huggingface.co/datasets/Francesco/truck-movement.

The CeTI-Age-Kinematics dataset provides full-body movements from 30 common daily tasks in nine categories, e.g., targeted reaching, lifting and walking tasks, and of multiple object interactions. Kinematic data were recorded at Dresden University of Technology (Germany) with informed consent of the participants under well-controlled conditions, covering multiple motion repetitions and task variations. The tasks were performed by 32 participants covering a wider age range, including older adults (66-75 years) and younger adults (19-28 years). Data were recorded using sensor suits and gloves with inertial measurement units (Rokoko Electronics, Denmark), with utilizing 33 sensors in total for full-body, wrist, and finger movements. The dataset also entails anthropometric body measurements and further demographic data. Additionally, the dataset provides spatial measurements of the experimental setups to enhance the interpretation of the kinematic data in relation to body characteristics and situational surroundings.The data records are structured according to the Motion-BIDS standard, with (i) meta-files describing the participants' characteristics such as demographic, anthropomorphic, and other data (see participant[.json|.tsv] files); (ii) metadata and additional information on the dataset (see dataset_description.json, README.md); (iii) TSV and BVH data that contain the kinematic motion data for each movement task, along with metadata on the task descriptions and instructions, (iv) source code to process and visualize the data, (v) materials that document the acquisition process such as recording protocols and exemplary videos of the motion executions.Detailed information, visualizations of the dataset content, and explanations of algorithmic approaches are available in the linked supplements, such as in the data descriptor and supplementary information accessible here: Pogrzeba, L., Muschter, E., Hanisch, S. et al. (2025). A Full-Body IMU-Based Motion Dataset of Daily Tasks by Older and Younger Adults. Scientific Data, 12, 531. https://doi.org/10.1038/s41597-025-04818-y