analiza aiqbench
AREA3D advances active 3D reconstruction by combining feed-forward models with vision-language guidance to autonomously select optimal viewpoints. This approach eliminates the need for expensive online optimization while achieving state-of-the-art accuracy in sparse-view scenarios, marking a significant step toward efficient autonomous scene understanding. The integration of semantic reasoning directly into geometric acquisition strategies represents a meaningful evolution in how agents interact with 3D environments.
AREA3D: Active Reconstruction Agent with Unified Feed-Forward 3D Perception and Vision-Language Guidance
arXiv:2512.05131v2 Announce Type: replace-cross Abstract: Active 3D reconstruction enables an agent to autonomously select viewpoints to efficiently obtain accurate and complete scene geometry, rather than passively reconstructing scenes from pre-collected images. However, existing active reconstruction methods often rely on hand-crafted geometric heuristics, which can lead to redundant observations without substantially improving reconstruction quality. To address this limitation, we propose AREA3D, an active reconstruction agent that leverages feed-forward 3D reconstruction models and vision-language guidance. Our framework decouples view-uncertainty modeling from the underlying feed-forward reconstructor, enabling precise uncertainty estimation without expensive online optimization. In addition, an integrated vision-language model provides high-level semantic guidance, encouraging informative and diverse viewpoints beyond purely geometric cues. Extensive experiments on both scene-level and object-level benchmarks demonstrate that AREA3D achieves state-of-the-art reconstruction accuracy, particularly in the sparse-view regime. Code will be made available at: https://github.com/TianlingXu/AREA3D .
Abstract:Active 3D reconstruction enables an agent to autonomously select viewpoints to efficiently obtain accurate and complete scene geometry, rather than passively reconstructing scenes from pre-collected images. However, existing active reconstruction methods often rely on hand-crafted geometric heuristics, which can lead to redundant observations without substantially improving reconstruction quality. To address this limitation, we propose AREA3D, an active reconstruction agent that leverages feed-forward 3D reconstruction models and vision-language guidance. Our framework decouples view-uncertainty modeling from the underlying feed-forward reconstructor, enabling precise uncertainty estimation without expensive online optimization. In addition, an integrated vision-language model provides high-level semantic guidance, encouraging informative and diverse viewpoints beyond purely geometric cues. Extensive experiments on both scene-level and object-level benchmarks demonstrate that AREA3D achieves state-of-the-art reconstruction accuracy, particularly in the sparse-view regime. Code will be made available at: this https URL .
Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Robotics (cs.RO)
arXiv:2512.05131 [cs.CV]
(or arXiv:2512.05131v2 [cs.CV] for this version)
Tianling Xu, Shengzhe Gan, Leslie Gu, Yuelei Li, Fangneng Zhan, Hanspeter Pfister. The IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) (The IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)), 2026
Text integral sub licența deschisă a publicației · 187 cuvinte
