{"ID":2848437,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2510.25233","arxiv_id":"2510.25233","title":"Hybrid Vision Servoing with Depp Alignment and GRU-Based Occlusion Recovery","abstract":"Vision-based control systems, such as image-based visual servoing (IBVS), have been extensively explored for precise robot manipulation. A persistent challenge, however, is maintaining robust target tracking under partial or full occlusions. Classical methods like Lucas-Kanade (LK) offer lightweight tracking but are fragile to occlusion and drift, while deep learning-based approaches often require continuous visibility and intensive computation. To address these gaps, we propose a hybrid visual tracking framework that bridges advanced perception with real-time servo control. First, a fast global template matcher constrains the pose search region; next, a deep-feature Lucas-Kanade module operating on early VGG layers refines alignment to sub-pixel accuracy (\u003c2px); then, a lightweight residual regressor corrects local misalignments caused by texture degradation or partial occlusion. When visual confidence falls below a threshold, a GRU-based predictor seamlessly extrapolates pose updates from recent motion history. Crucially, the pipeline's final outputs-translation, rotation, and scale deltas-are packaged as direct control signals for 30Hz image-based servo loops. Evaluated on handheld video sequences with up to 90% occlusion, our system sustains under 2px tracking error, demonstrating the robustness and low-latency precision essential for reliable real-world robot vision applications.","short_abstract":"Vision-based control systems, such as image-based visual servoing (IBVS), have been extensively explored for precise robot manipulation. A persistent challenge, however, is maintaining robust target tracking under partial or full occlusions. Classical methods like Lucas-Kanade (LK) offer lightweight tracking but are fr...","url_abs":"https://arxiv.org/abs/2510.25233","url_pdf":"https://arxiv.org/pdf/2510.25233v1","authors":"[\"Jee Won Lee\",\"Hansol Lim\",\"Sooyeun Yang\",\"Jongseong Brad Choi\"]","published":"2025-10-29T07:28:32Z","proceeding":"cs.RO","tasks":"[\"cs.RO\"]","methods":"[]","has_code":false}