AI Paper Atlas

Research map / Architectures and efficiency / RNNs and normalization

Accurate and Efficient Stereo Matching via Attention Concatenation Volume

Gangwei Xu, Yun Wang, Junda Cheng, Jinhui Tang, Xin Yan Yang

TPAMI 2023 · Architectures and efficiency › RNNs and normalization · approach: Dataset condensation and flow estimation · 130 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 attention weights to the concatenation volume (ablation)GwcNet · Scene FlowEPE (px): 0.76 -> 0.57Improved
add hourglass for attention weights generation (ablation)Gwc-att · Scene FlowEPE (px): 0.57 -> 0.52Improved
add supervision for attention weights (ablation)Gwc-att-hg · Scene FlowEPE (px): 0.52 -> 0.46Improved
add VAP (variance-aware prediction) to Fast-ACVNet (ablation)Fast-ACVNet · Scene FlowD1 (%): 2.51 -> 2.32Improved
reduce F2I sampled candidates from 48 to 16 (ablation)Fast-ACVNet with VAP · Scene Flowruntime (ms): 50 -> 35Mixed
use lightweight backbone and aggregation network (ablation)ACVNet · Scene Flowruntime (ms): 200 -> 62Mixed
add attention filter in Fast-ACV with one hourglass (ablation)Fast-ACV · Scene FlowEPE (px): 0.83 -> 0.64Improved
reduce hourglass number from 3 to 0 in ACV (ablation)Gwc-acv · Scene FlowEPE (px): 0.46 -> 0.59Hurt
replace GwcNet cost volume with ACV (PSMNet)PSMNet · Scene FlowEPE (px): 0.88 -> 0.63Improved
replace GwcNet cost volume with ACVGwcNet · Scene FlowEPE (px): 0.76 -> 0.46Improved
use Fast-ACV instead of ACV on GwcNetGwcNet · Scene Flowruntime (ms): 200 -> 81Mixed
integrate VAP into PSMNet (ablation)PSMNet · Scene FlowEPE (px): 0.88 -> 0.71Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 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