WAFT-Stereo: Warping-Alone Field Transforms for Stereo Matching

Yihan Wang; Jia Deng

cs.CVMar 25, 2026

WAFT-Stereo: Warping-Alone Field Transforms for Stereo Matching

Authors: Yihan Wang, Jia Deng

Organizations: Department of Computer Science, Princeton University

Abstract

We introduce WAFT-Stereo, a simple and effective warping-based method for stereo matching. WAFT-Stereo demonstrates that cost volumes, a common design used in many leading methods, are not necessary for strong performance and can be replaced by warping with improved efficiency. WAFT-Stereo ranks first on ETH3D (BP-0.5), Middlebury (RMSE), and KITTI (all metrics), reducing the zero-shot error by 81% on ETH3D, while being 1.8-6.7x faster than competitive methods. Code and model weights are available at https://github.com/princeton-vl/WAFT-Stereo.

Explore similar work

WAVE-Stereo: Warp-Aligned Volume Encoding for Stereo Matching
Jul 15, 2026Zehan Liu, Yage He, Xianwu GongStereo Matching Stereo Vision
LiteMatch: Lightweight Zero-Shot Stereo Matching via Cost Volume Stabilization
Jun 30, 2026Md Raqib Khan, Santosh Kumar Vipparthi, Subrahmanyam MuralaStereo Matching Cross-View
Lite Any Stereo V2: Faster and Stronger Efficient Zero-Shot Stereo Matching
Jun 23, 2026Junpeng Jing, Ronglai Zuo, Zhelun Shen +5Stereo Matching Stereo Vision

Jul 15, 2026 · cs.CVJ/K move · Enter open · S save

WAVE-Stereo: Warp-Aligned Volume Encoding for Stereo Matching

Zehan Liu, Yage He, Xianwu Gong

School of Electronics and Control Engineering Chang’an University Xi’an, China

Existing iterative stereo matching methods primarily adopt two types of correspondence representation: explicit matching search via correlation volumes and local residual refinement via warped features, yet the two remain separately modeled. We propose WAVE-Stereo, built on a core insight: correlation volumes and feature warping provide complementary matching cues. \textbf{GeoWarp Correspondence Encoder (GWCE)} encodes matching search, residual alignment, and disparity prior in parallel at the ConvGRU input. To mitigate matching degradation in textureless regions, we propose \textbf{Periodic Global Context Propagation (PGCP)}, which propagates global spatial information in a periodic manner. On five real-world benchmarks -- Middlebury, ETH3D, KITTI 2012, KITTI 2015, and Booster -- WAVE-Stereo achieves competitive zero-shot generalization accuracy without any external foundation model prior, achieving 3.18% D1-all on KITTI 2015, 4.42% Bad-2.0 on Booster, and 66ms real-time inference, striking a favorable balance between accuracy and efficiency. Our code is available at https://github.com/yamanoko-do/WAVE-Stereo.

Stereo Matching Stereo Vision Discrete Wavelet Transform Global Spatial Information Encoders Contextual Cues

Open paper Ask about it PDF

Abstract

Explore similar work

WAVE-Stereo: Warp-Aligned Volume Encoding for Stereo Matching

LiteMatch: Lightweight Zero-Shot Stereo Matching via Cost Volume Stabilization

Lite Any Stereo V2: Faster and Stronger Efficient Zero-Shot Stereo Matching