VlSomers/bpbreid
[WACV23] A strong baseline for body part-based person re-identification
This project helps security analysts or surveillance operators accurately identify individuals across different camera feeds, even when parts of their body are hidden or obstructed. It takes video frames or images of people as input and outputs a confidence score indicating the likelihood that two images show the same person, focusing on visible body parts rather than the whole body. This is designed for professionals who need to track individuals in complex, real-world environments with imperfect camera views.
250 stars.
Use this if you need to reliably match individuals captured by different surveillance cameras, especially when they might be partially obscured by obstacles or crowds.
Not ideal if your primary goal is general object detection or if you only need to identify people in completely clear, unobstructed views.
Stars
250
Forks
19
Language
Python
License
—
Category
Last pushed
Nov 05, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/computer-vision/VlSomers/bpbreid"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
ShuangLI59/person_search
Joint Detection and Identification Feature Learning for Person Search
VlSomers/keypoint_promptable_reidentification
[ECCV24] Keypoint Promptable Re-Identification: SOTA ReID method robust to occlusions and...
skylab-tech/ffhqr-dataset
FFHQR -- the first large-scale retouching dataset for computer vision research.
layumi/Vehicle_reID-Collection
:red_car: the collection of vehicle re-ID papers, datasets. :red_car:
tana0101/Hip-Joint-Keypoint-Detection
A research-oriented deep learning pipeline for automatic hip joint keypoint detection,...