Metraipolis › archive › dc_43eff13e4870
EgoSIS: From Factorized Visual Ego-Transitions to Motion-Canonical Spatial Evidence for UAV Reasoning — arXiv:2609.08938
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.08938
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-16T15:12:28.755Z
- SHA-256 of the archived text
f5e38077b2da64d88f1dcdbb779eda72cd71b83d3b4c81ed2899e32318996a69- Stable URL of this page
https://metraipolis.com/documento/dc_43eff13e4870
How to cite
EgoSIS: From Factorized Visual Ego-Transitions to Motion-Canonical Spatial Evidence for UAV Reasoning — arXiv:2609.08938 — arXiv (en), https://arxiv.org/abs/2609.08938 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_43eff13e4870 (Metraipolis archive dc_43eff13e4870, sha256 f5e38077b2da64d88f1dcdbb779eda72cd71b83d3b4c81ed2899e32318996a69)
BibTeX
@misc{dc43eff13e4870,
title = {EgoSIS: From Factorized Visual Ego-Transitions to Motion-Canonical Spatial Evidence for UAV Reasoning — arXiv:2609.08938},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.08938},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_43eff13e4870; sha256 f5e38077b2da64d88f1dcdbb779eda72cd71b83d3b4c81ed2899e32318996a69},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.08938. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
UAV video question answering requires separating camera motion from changes in the scene, but RGB-only multimodal models receive no explicit, stable reference for that separation. We present EgoSIS, a pose-free adapter that converts RGB-derived bidirectional flow into motion-canonical visual evidence in three stages. Factorized Visual Ego-Transitions (FVET) fits a robust image-plane transition and exposes motion, residual-support, and reliability factors. Reliability-Gated Ego-Transition Memory (ReTEM) uses reliability-weighted updates for a bounded history and re-anchors it at cuts or sustained uncertainty. Ego-Aligned Spatial Evidence (EASE) warps supported visual features into each segment's local anchor and injects four spatial evidence tokens per visual slice through zero-initialized residuals, without changing Qwen's visual-token count. On SIS-Bench, EgoSIS-8B obtains 89.9\% perception, 82.5\% perception-plus-memory, and 76.2\% overall accuracy, with the largest gains concentrated in self-awareness perception and memory. The adapter thus provides an interpretable interface between optical flow and spatial reasoning. Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-09 in cs.CV). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV). Pagina del lavoro: https://arxiv.org/abs/2609.08938 — 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 2026-10-16T14:49:57.369Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-09?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).