🧪 Test?View on arXiv
TraCE: Trajectory-Aware Concepts for Explainable Video Understanding
Wooil Lee, Author 2, Author 3, Author 4, Author 5
explainable AIvideo understandingmotion analysisconcept-based learning
2611.07009
Builder Relevance
3h ago80%
Abstract
This paper introduces TraCE, a framework that enhances video understanding by providing source-flexible motion concepts through clustering optical-flow trajectories.
Reality Card
Core Claim
TraCE achieves competitive task performance in action recognition and video question answering while improving interpretability of motion sources in video understanding.
Method / Result
Maintains an average accuracy drop of only 1.0 points for action recognition and a nominal gain of 0.4 points for VQA across six benchmarks.
Limitations
The reliance on specific benchmarks may limit generalizability and reproducibility across diverse video datasets.
Paper to code
Verified implementation resources so builders can test the paper’s claims instead of stopping at the abstract.
No verified implementation link has been attached yet. AIBuzzHub will keep this panel separate from unverified search results.