lorenzovaquero/BUSCA

[ECCV 2024] BUSCA: "Lost and Found: Overcoming Detector Failures in Online Multi-Object Tracking"

41
/ 100
Emerging

BUSCA helps surveillance operators, sports analysts, or autonomous vehicle developers maintain continuous tracking of multiple objects even when a detection system temporarily loses sight of them, for example, due to obstructions. It takes live video feeds as input and outputs a more reliable and consistent stream of identified object trajectories over time. This is for anyone who needs to track moving objects in real-time.

No commits in the last 6 months.

Use this if your multi-object tracking system frequently loses track of targets that are temporarily obscured or missed by its detection component, and you need to improve tracking persistence without re-training your entire system.

Not ideal if you are looking for a standalone object detection system or if your primary problem is inaccurate initial object identification rather than maintaining existing tracks.

multi-object tracking video surveillance sports analytics autonomous vehicles real-time monitoring
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 8 / 25
Maturity 16 / 25
Community 17 / 25

How are scores calculated?

Stars

43

Forks

9

Language

Python

License

GPL-3.0

Last pushed

Dec 06, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/computer-vision/lorenzovaquero/BUSCA"

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