AI Paper Atlas

Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning

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

Yihan Wang, Jia Deng

ECCV 2026 · Label-efficient and robust learning › Zero-/few-shot and meta-learning · approach: Transductive and semantic-dictionary zero-shot · 3 citations (OpenAlex)

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
add classification step before regression (vs regression only) (ablation)WAFT-Stereo · ETH3D / Middlebury-Q trainBP-1-all / BP-2-all: 2.59 -> 1.48 ETH3D; 6.45 -> 4.62 Middlebury-QImproved
add regression refinement after classification (vs classification only) (ablation)WAFT-Stereo · ETH3D / Middlebury-Q trainBP-1-all: 66.8 -> 1.48Improved
use cost volume for the classification step (ablation)WAFT-Stereo · ETH3D / Middlebury-Q trainBP-1-all: 1.48 -> 1.53No clear effect
use 40 bins instead of 5 (ablation)WAFT-Stereo · ETH3D trainBP-1-all: 2.46 -> 1.48Improved
use 80 bins instead of 40 (ablation)WAFT-Stereo · ETH3D / Middlebury-Q trainBP-1-all / BP-2-all: BP-1-all 1.48 -> 1.27; BP-2-all 4.62 -> 5.39Mixed
add high-resolution ResNet blocks (ablation)WAFT-Stereo · ETH3D trainBP-1-all: 1.99 -> 1.48Improved
use MoL loss instead of L1 (ablation)WAFT-Stereo · ETH3D trainBP-1-all: 2.44 -> 1.48Improved
scale backbone from DAv2-S to DAv2-L (ablation)WAFT-Stereo · ETH3D trainBP-1-all: 1.48 -> 0.80Improved
use DINOv3-L instead of DAv2-L (ablation)WAFT-Stereo · ETH3D trainBP-1-all: 0.80 -> 0.87No clear effect
train on SceneFlow only instead of 7.7M synthetic pairs (ablation)WAFT-Stereo (DAv2-S) · ETH3D trainBP-1-all: 0.86 -> 71.0Hurt
replace cost volumes with warping (full model) vs FoundationStereoWAFT-Stereo · ETH3D test zero-shotBP-0.5-noc: 2.31 -> 0.89Improved
WAFT-Stereo vs S2M2-XL on Middlebury BP-2WAFT-Stereo · Middlebury testBP-2-noc: 1.15 vs 2.53Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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