AI Paper Atlas

Research map / 3D vision / Depth and stereo

Track Everything Everywhere Fast and Robustly

Yunzhou Song, Jiahui Lei, Ziyun Wang, Lingjie Liu, Kostas Daniilidis

ECCV 2024 · 3D vision › Depth and stereo · approach: Learned optical and scene flow · 11 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
replace photometric neural-rendering geometry with optimizable monocular metric depth maps (ablation)CaDeX++ tracker · DAVIS subsetAJ: 42.0 -> 48.6Improved
add DINOv2 long-term correspondence supervision (ablation)CaDeX++ tracker · DAVIS subsetAJ: 45.6 -> 48.6Improved
use CaDeX++ local invertible mapping instead of global MLP NVP (ablation)CaDeX++ tracker · DAVIS subsetAJ: 48.2 -> 48.6No clear effect
use CaDeX++ instead of global NVP (ablation)CaDeX++ tracker · DAVIS subsetTC: 0.97 -> 1.14Hurt
add depth prior to stabilize optimization across random seeds (ablation)CaDeX++ tracker · motocross-jump delta_avg over seedsstd: 23.5 -> 0.5Improved
replace OmniMotion with CaDeX++ and depth priorOmniMotion · TAP-Vid DAVISAJ: 51.7 -> 59.4Improved
replace OmniMotion with CaDeX++ on synthetic RGB-StackingOmniMotion · TAP-Vid RGB-StackingAJ: 77.5 -> 75.4Hurt
replace OmniMotion with CaDeX++ in training timeOmniMotion · DAVISoptimization timeImproved

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