Metraipolis › archive › dc_b86e46aa7345
TraCE: Trajectory-Aware Concepts for Explainable Video Understanding — arXiv:2611.07009
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.07009
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-06T14:53:15.791Z
- SHA-256 of the archived text
7e28ee4a623de2d4178b7c8fecf8f925e4653311aa7fe923047cbeacb19ee1f3- Stable URL of this page
https://metraipolis.com/documento/dc_b86e46aa7345
How to cite
TraCE: Trajectory-Aware Concepts for Explainable Video Understanding — arXiv:2611.07009 — arXiv (en), https://arxiv.org/abs/2611.07009 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_b86e46aa7345 (Metraipolis archive dc_b86e46aa7345, sha256 7e28ee4a623de2d4178b7c8fecf8f925e4653311aa7fe923047cbeacb19ee1f3)
BibTeX
@misc{dcb86e46aa7345,
title = {TraCE: Trajectory-Aware Concepts for Explainable Video Understanding — arXiv:2611.07009},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.07009},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_b86e46aa7345; sha256 7e28ee4a623de2d4178b7c8fecf8f925e4653311aa7fe923047cbeacb19ee1f3},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.07009. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Video understanding models often rely on motion, yet their explanations can obscure which motion sources support a prediction. We identify motion-source under-specification as a limitation of existing video XAI: region-centric explanations can be source-ambiguous, while motion vocabularies restricted to single-person pose sequences are source-incomplete. We propose TraCE (Trajectory-Aware Concepts for Explainable Video Understanding), a concept-based framework that discovers source-flexible motion concepts by clustering optical-flow trajectories from training videos. These concepts represent dynamic evidence from people, objects, and interactions, while object and scene concepts account for static context. The resulting typed trajectory/object/scene interface supports class prediction for action recognition and question-conditioned concept matching for video question answering (VQA). We evaluate TraCE across five random seeds on six benchmarks: NTU RGB+D Mutual, Chiral SSv2, UCF-101, KTH, TOMATO, and STAR. Relative to non-interpretable baselines, TraCE maintains competitive task performance, with an average accuracy drop of 1.0 points for action recognition and a nominal gain of 0.4 points for VQA. Motion perturbation and concept-removal tests provide complementary evidence of directional semantic alignment and decision faithfulness, while sample-level interventions illustrate targeted model debugging. Our project page is available at this https URL . Tema degli esploratori: statistica e scienza dei dati (elenco arXiv cs.CV, nuovi lavori del giorno). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV). Pagina del lavoro: https://arxiv.org/abs/2611.07009 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-11-06T14:25:42.603Z dalla pagina https://arxiv.org/list/cs.CV/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).