Metraipolis › archive › dc_6888cc21facd

Fine-Grained Action Recognition with Cross-Attentive Latent Sparse Experts — arXiv:2608.13458

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.13458
License (as archived)
CC0-1.0
Archived
2026-10-19T04:55:56.483Z
SHA-256 of the archived text
facce62c4b123c57f0384ba4a90652598ffd02e6da45c0bf60fc2a11c031a6d9
Stable URL of this page
https://metraipolis.com/documento/dc_6888cc21facd

How to cite

Fine-Grained Action Recognition with Cross-Attentive Latent Sparse Experts — arXiv:2608.13458 — arXiv (en), https://arxiv.org/abs/2608.13458 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_6888cc21facd (Metraipolis archive dc_6888cc21facd, sha256 facce62c4b123c57f0384ba4a90652598ffd02e6da45c0bf60fc2a11c031a6d9)
BibTeX
@misc{dc6888cc21facd,
  title = {Fine-Grained Action Recognition with Cross-Attentive Latent Sparse Experts — arXiv:2608.13458},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.13458},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_6888cc21facd; sha256 facce62c4b123c57f0384ba4a90652598ffd02e6da45c0bf60fc2a11c031a6d9},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.13458. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Fine-grained human action recognition (FHAR) must distinguish visually similar actions that differ mainly in body configuration, timing, or local appearance. RGB representations retain visual context but often suppress joint-level geometry, whereas skeleton representations encode kinematics but discard dense spatial detail. We introduce FineX, which factorizes fine-grained cues into RGB appearance, pose heatmap geometry, and skeletal-graph topology. Pairwise cross-attention enables symmetric, stream-preserving information exchange, followed by a streamwise latent sparse Mixture-of-Experts that routes each representation to a content-dependent subset of shared experts, regularized by a load-balancing objective. FineX achieves state-of-the-art results on Gym99, Gym288, and Diving48. On the long-tailed Gym288, it raises mean class accuracy from 68.6% to 76.2% (+7.6 points) without textual supervision or large-scale vision-language pre-training, demonstrating the benefit of structured visual-pose-graph fusion and conditional expert refinement for FHAR.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-08-14 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2608.13458 — 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 2026-10-19T04:46:30.248Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-08-14?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).