gloriacao/Datachallenge_CM4606

Dataset Card for Falls and Non-Falls Dataset

Dataset consisting of 4,002 images of adults and elderly people falling, which were all reorganised into two categories: "fall" and "non-fall".

Dataset Details

The dataset was created for a project in the Unboxing the Algorithms course at Erasmus University Rotterdam, supervised by Professor JF Ferreira Gonçalves. The project aims to develop and evaluate an image-based machine learning model that can identify images as either fall-related or non-fall-related in order to detect an emergency when necessary.

Dataset Description

The dataset contains 3,998 images from multiple publicly available fall-detection datasets, as well as additional manually curated images. The goal was to create a diverse dataset that could distinguish between falls and non-falls in environments such as homes and residential care facilities. 10% of the total images were allocated to a validation set. The goal of this dataset is to serve the project's purpose of training machine learning models to identify emergency situations involving fallen people.

• Curated by: Kerija Broka, Faye Dolan, Wai Hou Foong, Gloria Caoduro

• License: Creative Commons Attribution 4.0 (CC-BY-4.0)

Uses

Direct Use

The purpose of the dataset is to train a machine learning image classifier to detect potential falls by elderly individuals by classifying images as either "fall" or "non-fall."

Out-of-Scope Use

This dataset is not intended for applications such as biometric identification, facial recognition, medical diagnosis, or decision-making. It is also not intended for real-world fall detection systems that rely solely on camera-based monitoring because it may require multimodal sensing to improve reliability.

Dataset Structure

• Images: 4,002

• Labels: "Fall", "Non-Fall"

• Demographic indicators: elderly people and partly all ages, different skin tones, hair color, clothing styles and body shape.

• Dataset split: training, validation.

Curation Rationale

The dataset was created to improve the ability of machine learning to detect falls in elderly people. It includes subjects with different characteristics and skin colors in an attempt to minimize biases.

Source Data

Data Collection and Processing

The dataset was developed using publicly available image and video datasets, from which frames were extracted. Over 400 images were manually selected from Unsplash.com and Pexels.com.

Personal and Sensitive Information

Although the dataset does not contain personal identifiers, it includes images of individuals in potentially vulnerable situations. Therefore, the dataset may contain implicitly sensitive information, such as visible physical characteristics.

Bias, Risks, and Limitations

In order to minimize biases regarding age, sex, race, and clothing, individuals with different characteristics were included. This was done to reduce the risk of the model working only on specific types of individuals.

Recommendations

Users of this dataset should be aware of its biases and limitations.

Citation

BibTeX:

@dataset{alam2024gmdcsa24, author={Alam, Ekram}, title={GMDCSA24: A Dataset for Human Fall Detection in Videos},year= {2024},version={2.0},publisher={Zenodo},doi= {10.5281/zenodo.12921216},url= {https://doi.org/10.5281/zenodo.12921216}}

@dataset{brooka2026elderly, author={Brooka, Kerija}, title={Manually Curated Dataset of Elderly Individuals in Indoor Activities}, year={2026}, note={Images collected from Unsplash and Pexels}}

@dataset{charoensri_fallvideo, author={Charoensri, Payut}, title={Fall Video Dataset}, year= {n.d.}, publisher={Kaggle}, url= {https://www.kaggle.com/datasets/payutch/fall-video-dataset}}

@dataset{robinovitch2018falls, author={Robinovitch, Stephen}, title={Falls Experienced by Older Adult Residents in Long-Term Care Homes}, year={2018}, publisher={Databrary}, url={https://databrary.org/volume/739}, note={Accessed: 2026-03-09}}

@article{maldonado2019fallen, author={Maldonado-Bascón, Saturnino and Iglesias-Iglesias, Carlos and Martìn-Martìn, Pablo and Lafuente-Arroyo, Sergio}, title={Fallen People Detection Capabilities Using Assistive Robot}, journal={Electronics}, volume={8}, number={9}, pages={915}, year={2019}, doi={10.3390/electronics8090915}}

@dataset{lin2025interior, author = {Lin, Hunter}, title = {Interior}, year = {2025}, publisher = {Roboflow Universe}, url = {https://universe.roboflow.com/hunter-lin-mw9bp/interior-fb01j}}

@dataset{david2025roominterior, author = {David}, title = {Room-Interior Dataset}, year = {2025}, month = {April}, publisher = {Roboflow Universe}, url = {https://universe.roboflow.com/david-dtylq/room-interior-aa0wb}}

@dataset{pahari2025falldetection, author = {Pahari, Suman}, title = {FallDetection}, year = {2025}, month = {July}, publisher = {Roboflow Universe}, url = {https://universe.roboflow.com/suman-pahari-ieiix/falldetection-nb4dq}}

@dataset{elderlyfalldetection2025, author = {{Elderly Fall Detection}}, title = {Fall Detection}, year = {2025}, month = {July}, publisher = {Roboflow Universe}, url = {https://universe.roboflow.com/elderly-fall-detection/fall-detection-wvbiq}}

@dataset{finalcapstone2022fall, author = {{finalcapstone}}, title = {Fall-Detected}, year = {2022}, month = {September}, publisher = {Roboflow Universe}, url = {https://universe.roboflow.com/finalcapstone/fall-detected}}

APA:

Alam, E. (2024). GMDCSA24: A dataset for human fall detection in videos (Version 2.0) [Dataset]. Zenodo. https://doi.org/10.5281/zenodo.12921216

Brooka, K. (2026). Manually curated dataset of elderly individuals in indoor activities [Dataset]. Images collected from Unsplash and Pexels.

Charoensri, P. (n.d.). Fall video dataset [Dataset]. Kaggle. https://www.kaggle.com/datasets/payutch/fall-video-dataset

Robinovitch, S. (2018). Falls experienced by older adult residents in long-term care homes [Dataset]. Databrary. Retrieved March 9, 2026, from https://databrary.org/volume/739

Maldonado-Bascón, S., Iglesias-Iglesias, C., Martín-Martín, P., & Lafuente-Arroyo, S. (2019). Fallen people detection capabilities using assistive robot [Dataset associated with publication]. Electronics, 8(9), 915. https://doi.org/10.3390/electronics8090915

Lin, H. (2025). Interior [Dataset]. Roboflow Universe. https://universe.roboflow.com/hunter-lin-mw9bp/interior-fb01j

David. (2025, April). Room-interior dataset [Dataset]. Roboflow Universe. https://universe.roboflow.com/david-dtylq/room-interior-aa0wb

Pahari, S. (2025, July). FallDetection [Dataset]. Roboflow Universe. https://universe.roboflow.com/suman-pahari-ieiix/falldetection-nb4dq

Elderly Fall Detection. (2025, July). Fall detection [Dataset]. Roboflow Universe. https://universe.roboflow.com/elderly-fall-detection/fall-detection-wvbiq

finalcapstone. (2022, September). Fall-detected [Dataset]. Roboflow Universe. https://universe.roboflow.com/finalcapstone/fall-detected