AI Paper Atlas

Research map / Architectures and efficiency / RNNs and normalization

Volumetric Correspondence Networks for Optical Flow

Gengshan Yang, Deva Ramanan

NeurIPS 2019 · Architectures and efficiency › RNNs and normalization · approach: Dataset condensation and flow estimation · 161 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
use true 4D convolution on cost volume instead of 2D DenseNet matching (ablation)single-stage matching module · KITTI-15 train 0-32pxEPE: 2.64 -> 2.30Improved
separate 4D kernels into spatial and WTA 2D kernels (ablation)Full-4D · KITTI-15 train 0-32pxEPE: 2.30 -> 2.31No clear effect
use volumetric U-Net instead of plain 4D conv stack (ablation)Ours-UNet · KITTI-15 train 0-32pxEPE: -0.02 EPE for plain, +20.9 GFlopsNo clear effect
add multi-channel cost volume (ablation)Ours-UNet · KITTI-15 train 0-32pxEPE: +0.32 when removedImproved
use truncated soft-argmin instead of soft-argmin (ablation)Ours-UNet · KITTI-15 train 0-32pxEPE: +0.10 when replacedImproved
use truncated soft-argmin instead of direct regression (ablation)Ours-UNet · KITTI-15 train 0-32pxEPE: +0.58 when replacedImproved
add out-of-range detection module (ablation)Ours-UNet · KITTI-15 train 0-32pxEPE: +0.07 when removedImproved
VCN vs PWC-Net when extending search range beyond training rangeflow network · KITTI-15 train 0-64pxEPE: 9.44 -> 3.27Improved
VCN vs best prior fine-tuned methodsoptical flow · Sintel test cleanEPE: 3.45 -> 2.81Improved
VCN vs PWC-Net on stereo without stereo trainingflow network · KITTI-15 stereo trainD1-all: 23.30 -> 8.73Improved
use 2D search window for stereo with imperfect rectificationstereo matching · Middlebury-14avgerr: 9.92 (PWC-Net) -> 8.79Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 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