AI Paper Atlas

Research map / 3D vision / Depth and stereo

GS-ASM: 2DGS-Supervised Active Stereo Matching

Zhengling Wu, Rongfeng Lu, Quan Chen, Longjian Zeng, Ming Lu, Yaoqi Sun, Yahong Chen, Baofeng Ji, Chenggang Yan

CVPR 2026 · 3D vision › Depth and stereo · approach: Foundation models for monocular geometry

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
train with 2DGS proxy-label supervision plus novel views and hybrid regularization instead of ActiveZero hybrid-domain trainingRAFT stereo network · Simulated test set (Blender)EPE (px): 0.4715 -> 0.1279Improved
train with GS-ASM instead of ActiveZero baselinePSMNet · Simulated test set (Blender)EPE (px): 0.4300 -> 0.2613Improved
train with GS-ASM instead of ActiveZero baselineStereoNet · Simulated test set (Blender)EPE (px): 0.5551 -> 0.4124Improved
add 2DGS proxy labels for supervised training (ablation)ActiveZero baseline on RAFT · Simulated test set (Blender)EPE (px): 0.4715 -> 0.3482Improved
add novel-viewpoint data augmentation from 2DGS (ablation)Baseline with proxy labels on RAFT · Simulated test set (Blender)EPE (px): 0.3482 -> 0.2736Improved
add hybrid supervision regularization (adaptive loss weights) (ablation)Proxy labels + novel viewpoints on RAFT · Simulated test set (Blender)EPE (px): 0.2736 -> 0.1279Improved
add hybrid supervision regularization (ablation)Proxy labels + novel viewpoints on RAFT · Simulated test set (Blender)5px accuracy: 0.9945 -> 0.9955Improved
fit 2DGS on IR images captured at higher emitter intensity instead of zero emitter power (ablation)active 2DGS · 2DGS reconstruction of IR imagesPSNR: 33.23 -> 31.75Hurt
use IR images at higher emitter intensity for binary pattern extraction (ablation)binary IR pattern extraction · 2DGS reconstruction of IR imagesLPIPS: 0.0762 -> 0.3936Hurt
use D435 depth as supervision instead of 2DGS proxy labelsRAFT · Simulated test set (Blender)EPE (px): 0.1279 (ours) vs 0.5428 (D435)Hurt
use ActiveZero hybrid-domain training in real scenesActiveZero · Real-world scenes (qualitative)depth qualityHurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy