Unity-Technologies/PeopleSansPeople

Unity's privacy-preserving human-centric synthetic data generator

42
/ 100
Emerging

PeopleSansPeople helps computer vision engineers develop models for human detection and pose estimation without needing extensive real-world human data. It takes descriptions of virtual scenes with human characters and outputs large, diverse synthetic datasets complete with 2D/3D bounding boxes, segmentation masks, and human keypoint labels. This is ideal for researchers and engineers building computer vision applications that interact with people.

321 stars. No commits in the last 6 months.

Use this if you need to train or pre-train computer vision models that analyze human actions or presence, but face challenges with data privacy, collection costs, or diversity in real-world datasets.

Not ideal if your computer vision task does not involve human subjects, or if you exclusively require real-world, non-synthetic data for training and validation.

computer-vision human-pose-estimation synthetic-data-generation privacy-preserving-AI AI-model-training
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 16 / 25

How are scores calculated?

Stars

321

Forks

34

Language

C#

License

Apache-2.0

Last pushed

Mar 05, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/Unity-Technologies/PeopleSansPeople"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.