Universe

webot fall Computer Vision Dataset

byheartbrokerTask:
Object Detection
License:CC BY 4.0

How to use the webot fall Detection API

Try This Model

Drop an image here or click to upload

Or try a test image 

Code Snippets

from inference_sdk import InferenceHTTPClient

CLIENT = InferenceHTTPClient(
    api_url="https://serverless.roboflow.com",
    api_key="API_KEY"
)

result = CLIENT.infer("YOUR_IMAGE.jpg", model_id="heartbroker/webot-fall-instant-1")
Give your agent everything it needs

Or, Use Free Person, 0 and 1 Detection API

Powered by general detection model

Code
pip install inference-sdk
# 1. Import the library
from inference_sdk import InferenceHTTPClient

# 2. Connect to your workspace
client = InferenceHTTPClient(
  api_url="https://serverless.roboflow.com",
  api_key="API_KEY"
)

# 3. Run your workflow on an image
result = client.run_workflow(
  workspace_name="<YOUR_WORKSPACE>",
  workflow_id="<YOUR_WORKFLOW_ID>",
  images={
    "image": "YOUR_IMAGE.jpg"  # Path to your image file
  },
  parameters={
    "classes": "person, 0, 1"
  },
  use_cache=True  # cache workflow definition for 15 minutes
)

# 4. Get your results
print(result)

Run on custom image

Drop an image here or click to upload

Detecting classes:
Or try a test image 
See all 253 imageswebot fall person sample showing 0webot fall person sample showing 0webot fall person sample showing 1webot fall person sample showing 0webot fall person sample showing 1webot fall person sample showing 0webot fall person samplewebot fall person sample showing 0webot fall person sample showing 1webot fall person sample showing 1webot fall person sample showing 1webot fall person sample showing 0

About webot fall Model

Description: This dataset is collected from a Webots robot simulation environment for human fall detection research. All images are captured from a fixed overhead (bird's eye view) camera mounted approximately 3.5 meters above a simulated indoor scene, covering a 9.9m × 6.6m residential environment including living room, dining area, kitchen, and study zones. The humanoid pedestrian model is configured with 5 preset fall poses (right side fall, left side fall, face-down, face-up, half-fall) alongside normal standing and walking states. This dataset is specifically designed to complement real-world fall datasets by providing overhead-angle samples that are typically underrepresented in publicly available fall detection datasets. All images are annotated with bounding boxes in YOLO format for two classes:

fall — person in a fallen state stand — person in normal standing or walking state

Key Features:

Simulated indoor residential scene with realistic furniture layout (sofa, dining table, kitchen appliances, bookshelf, etc.) Consistent overhead camera angle (no perspective distortion from side views) Clean background with stable lighting conditions Suitable for training or fine-tuning YOLO-series models Complements the Multicam Fall Dataset for multi-source training

Intended Use: This dataset is intended for researchers working on vision-based fall detection systems, elderly care robotics, and intelligent surveillance applications. It was collected as part of an undergraduate thesis project integrating YOLOv8 detection with PR2 robot autonomous rescue navigation in Webots simulation. Citation: If you use this dataset in your research, please consider citing: Webots Fall Detection Dataset (Bird's Eye View) Collected via Webots R2023b simulation platform Part of: Machine Vision-Based Fall Detection and Rescue Virtual System Undergraduate Thesis, 2026

Cite This Project

LicenseCC BY 4.0

If you use this dataset in a research paper, please cite it using the following BibTeX:

@misc{ webot-fall_dataset,
  title = { webot fall Dataset },
  type = { Open Source Dataset },
  author = { heartbroker },
  howpublished = { \url{ https://universe.roboflow.com/heartbroker/webot-fall } },
  url = { https://universe.roboflow.com/heartbroker/webot-fall },
  journal = { Roboflow Universe },
  publisher = { Roboflow },
  year = { 2026 },
  month = { mar },
  note = { visited on 2026-07-29 },
}

Similar Projects

See More