EgoSIS: From Factorized Visual Ego-Transitions to Motion-Canonical Spatial Evidence for UAV Reasoning

• URL: http://arxiv.org/abs/2609.08938v2
• Date: Wed, 09 Sep 2026 04:55:33 GMT
• Title: EgoSIS: From Factorized Visual Ego-Transitions to Motion-Canonical Spatial Evidence for UAV Reasoning
• Abstract summary: EgoSIS is a pose-free adapter that converts RGB-derived bidirectional flow into motion-canonical visual evidence.<n>On SIS-Bench, EgoSIS-8B obtains 89.9% perception, 82.5% perception-plus-memory, and 76.2% overall accuracy.
• Score: 4.410395389325955
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: 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.

Related papers

• sensVLA: Spatially-Grounded Vision-Language-Action Model for Autonomous Wheel Loader [0.0]
  We present sensVLA, a Vision-Language-Action architecture for autonomous wheel-loader control.<n>It combines a Qwen3-2B Vision-Language Model (VLM) with a fully trainable transformer action expert trained by flow-matching velocity regression.<n>sensVLA routes Bird's-Eye-View (BEV) features, extracted from fused front and rear lidar, directly to the action expert through a dedicated cross-attention pathway.
  arXiv  Detail & Related papers  (2026-09-15T11:29:50Z)
• RASA: Disentangled Spatial-Motional Priors for Cross-Identity Character Animation [85.41886214825894]
  Cross-identity character animation aims to drive a target identity from a reference image to follow the motion of a source character from a driving video.<n>We introduce Reference-Aware Structural Alignment (RASA), a framework that disentangles spatial mapping from motion control.<n>Our work establishes a new paradigm showing disentangled spatial and motional priors are key to robust character animation.
  arXiv  Detail & Related papers  (2026-08-28T11:36:44Z)
• VADER: Adaptive Debiasing for Hallucination Mitigation in Video Large Language Models [30.792102059830274]
  Video-Adaptive Debiasing via Evidence Reweighting (VADER) is a training-free framework with two complementary modules.<n>Visual Focus Reallocation (VFR) automatically instantiates an intervention policy for each video-question input.<n> Selective Evidence Erasure (SEE) independently masks high-importance visual tokens in every frame.
  arXiv  Detail & Related papers  (2026-08-09T10:09:11Z)
• Difference-Based Relational Learning for Zero-Shot Object-Goal Visual Navigation With Direct Sim-to-Real Transfer [0.5371337604556311]
  Temporal Difference-Relational Network (T-DRN) for robust zero-shot sim-to-real transfer.<n>Experiments in AI2-THOR demonstrate that T-DRN improves zero-shot generalization in terms of success rates over strong baselines.<n>T-DRN is systematically validated on a physical wheeled robot, demonstrating robust performance under real sensing and actuation constraints.
  arXiv  Detail & Related papers  (2026-07-17T05:35:40Z)
• REMIND: RE-Identification with Memory for INDoor Navigation [39.146761527401424]
  Online tracker designed for long-term multi-object re-identification of generic indoor objects from monocular RGB imagery.<n>Motivated by evidence from visual cognition that humans rely on accumulated appearance familiarity and spatial context rather than explicit self-localization.<n>On ScanNet++, it attains the highest IDF1 in every setting but one, end-to-end detection over all scenes.
  arXiv  Detail & Related papers  (2026-07-10T10:30:50Z)
• Persistent Visual Memory: Sustaining Perception for Deep Generation in LVLMs [66.15429821530503]
  Persistent Visual Memory is a lightweight learnable module designed to strengthen sustained, on-demand access to visual evidence.<n>Experiments on Qwen3-VL models demonstrate that PVM brings notable improvements with negligible parameter overhead.<n>In-depth analysis reveals that PVM shows improved robustness in longer generations and accelerates internal prediction convergence.
  arXiv  Detail & Related papers  (2026-05-01T17:54:37Z)
• TransVLM: A Vision-Language Framework and Benchmark for Detecting Any Shot Transitions [52.61846373082384]
  Shot Transition Detection (STD) inherently struggles with complex transitions.<n>We propose TransVLM, a Vision-Language Model (VLM) framework for STD.<n>Extensive experiments demonstrate that TransVLM achieves superior overall performance.
  arXiv  Detail & Related papers  (2026-04-30T15:05:06Z)
• Micro-Expression-Aware Avatar Fingerprinting via Inter-Frame Feature Differencing [12.334126578226318]
  Avatar fingerprinting verifies who drives a synthetic talking-head video rather than whether it is real.<n>Existing methods rely on a fixed, non-differentiable landmark extraction stage.<n>We propose a preprocessing-free system built on a micro-expression-aware backbone operating on raw video frames.
  arXiv  Detail & Related papers  (2026-04-25T10:57:39Z)
• Robust Egocentric Referring Video Object Segmentation via Dual-Modal Causal Intervention [58.05340906967343]
  Egocentric Referring Video Object (Ego-RVOS) aims to segment the specific object actively involved in a human action, as described by a language query, within first-person videos.<n>Existing methods often struggle, learning spurious correlations from skewed object-action pairings in datasets.<n>We introduce Causal-REferring (CERES), a plug-in causal framework that adapts strong, pre-trained RVOSs to the egocentric domain.
  arXiv  Detail & Related papers  (2025-12-30T16:22:14Z)
• Latent Diffusion Model without Variational Autoencoder [78.34722551463223]
  SVG is a novel latent diffusion model without variational autoencoders for visual generation.<n>It constructs a feature space with clear semantic discriminability by leveraging frozen DINO features.<n>It enables accelerated diffusion training, supports few-step sampling, and improves generative quality.
  arXiv  Detail & Related papers  (2025-10-17T04:17:44Z)
• EgoPrune: Efficient Token Pruning for Egomotion Video Reasoning in Embodied Agent [41.11532785015233]
  EgoPrune is a training-free token pruning method tailored for egomotion video reasoning.<n>EgoPrune consistently outperforms prior training-free methods across various pruning ratios.<n>We deploy EgoPrune on an embodied agent equipped with a Jetson Orin NX 16GB edge device.
  arXiv  Detail & Related papers  (2025-07-21T09:27:45Z)
• SD-ReID: View-aware Stable Diffusion for Aerial-Ground Person Re-Identification [74.36139886192495]
  We propose a novel generative framework named SD-ReID for AG-ReID.<n>We first train a ViT-based model to extract person representations along with controllable conditions, including identity and view conditions.<n>We then fine-tune the Stable Diffusion (SD) model to enhance person representations guided by these controllable conditions.
  arXiv  Detail & Related papers  (2025-04-13T12:44:50Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.