arXiv cs.AI· Qi Xiong, Qinghua Zhao, Yongqiang Xi·· 14 小时前AI 评分18
E3Stereo：用学习到的几何边缘改进深度边界处的立体匹配
E3tereo: Improving Stereo Matching at Depth Boundaries with Learned Geometric Edges
AI 导读
E3Stereo 从视差监督中学习几何边缘，并结合主干视差与图像细节预测全分辨率视差修正，同时用同一边缘图在训练中强化边界误差。在 Scene Flow 数据集上，该方法使 RAFT-Stereo、IGEV 和 MonSter 的深度边界 EPE 降低 9.5%-14.3%，整体 EPE 也同步下降。
正文 · 原文
Abstract:Depth boundaries remain a major source of error in stereo matching. For IGEV on Scene Flow, geometric edge pixels account for only 6% of evaluated pixels but about half of the total absolute disparity error. Photometric edges can respond to texture changes without corresponding depth discontinuities, making them unreliable cues for boundary refinement. We propose E3Stereo, which learns geometric edges from disparity supervision and combines these boundary cues with the backbone disparity and image details to predict full-resolution disparity corrections. The same edge map also emphasizes boundary errors during training, while joint optimization allows the edge predictor to adapt to disparity refinement. On the Scene Flow dataset, E3Stereo reduces end-point error (EPE) at depth boundaries by 9.5%-14.3% across RAFT-Stereo, IGEV, and MonSter, while also reducing overall EPE. After fine-tuning, our IGEV-based model improves upon IGEV on all eleven reported metrics across ETH3D, KITTI 2012, and KITTI 2015. The source code will be made publicly available upon acceptance.
| Comments: | submitted to cvpr |
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.33704 [cs.CV] |
| (or arXiv:2610.33704v1 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2610.33704 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Qinghua Zhao [view email]
[v1]
Sat, 24 Oct 2026 14:35:22 UTC (6,799 KB)
来源：arXiv cs.AI · arxiv.org