AREA3D: Active Reconstruction Agent with Unified Feed-Forward 3D Perception and Vision-Language Guidance

Tianling Xu1,2 Shengzhe Gan1 Leslie Gu2 Yuelei Li3 Fangneng Zhan4 Hanspeter Pfister2
1 Southern University of Science and Technology   2 Harvard University
3 California Institute of Technology   4 MIT
CVPR 2026
SUSTech Harvard Caltech MIT
Paper Code arXiv

Overview

AREA3D Teaser

AREA3D actively selects viewpoints to achieve accurate 3D reconstruction with vision-language guidance.

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.

Citation

@inproceedings{xu2026area3d,
  title={AREA3D: Active Reconstruction Agent with Unified Feed-Forward 3D Perception and Vision-Language Guidance},
  author={Xu, Tianling and Gan, Shengzhe and Gu, Leslie and Li, Yuelei and Zhan, Fangneng and Pfister, Hanspeter},
  booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
  year={2026}
}