AI Paper Atlas

Research map / 3D vision / Depth and stereo

Selection and Cross Similarity for Event-Image Deep Stereo

Hoonhee Cho, Kuk-Jin Yoon

ECCV 2022 · 3D vision › Depth and stereo · approach: Learned stereo with specialized sensors · 35 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 differentiable event selection (DES) (ablation)GwcNet-g baseline · MVSEC Split 1mean disparity error: 0.4020 -> 0.4111Hurt
add differentiable event selection (DES) (ablation)GwcNet-g baseline · MVSEC Split 3mean disparity error: 0.4358 -> 0.4153Improved
add differentiable event selection (DES) (ablation)GwcNet-g baseline · MVSEC Split 3one-pixel accuracy: 91.7874 -> 93.7863Improved
add differentiable event selection (DES) (ablation)GwcNet-g baseline · MVSEC Split 3mean depth error: 16.5391 -> 14.3000Improved
add neighbor cross similarity feature (NCSF) (ablation)GwcNet-g baseline · MVSEC Split 1mean disparity error: 0.4020 -> 0.3909Improved
add neighbor cross similarity feature (NCSF) (ablation)GwcNet-g baseline · MVSEC Split 3one-pixel accuracy: 91.7874 -> 93.0166Improved
add NCSF on top of DES (ablation)GwcNet-g + DES · MVSEC Split 3mean disparity error: 0.4153 -> 0.3895Improved
add DES on top of NCSF (ablation)GwcNet-g + NCSF · MVSEC Split 1mean disparity error: 0.3909 -> 0.3776Improved
select K=3 of B=5 event bins instead of K=1 (ablation)Proposed · MVSEC Split 1one-pixel accuracy: 91.6541 -> 94.7201Improved
select K=3 of B=5 event bins instead of all 5 (ablation)Proposed · MVSEC Split 1one-pixel accuracy: 94.1027 -> 94.7201Improved
select K=3 instead of K=2 (ablation)Proposed · MVSEC Split 1one-pixel accuracy: 93.2581 -> 94.7201Improved
full method vs EIS event-image fusionProposed · MVSEC Split 3one-pixel accuracy: 88.1 -> 94.0Improved

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