AI Paper Atlas

Research map / 3D vision / Depth and stereo

C4D: 4D Made from 3D through Dual Correspondences

Shizun Wang, Zhenxiang Jiang, Xingyi Yang, Xinchao Wang

ICCV 2025 · 3D vision › Depth and stereo · approach: Feed-forward 3D reconstruction · 7 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 global alignment with C4D optimization on DUSt3R weightsDUSt3R · Sintel camera poseRPE rot: 18.038 -> 0.948Improved
replace global alignment with C4D on MonSt3R weightsMonSt3R · Sintel camera poseATE: 0.158 -> 0.103Improved
replace global alignment with C4D on MASt3R weightsMASt3R · Sintel camera poseATE: 0.437 -> 0.448 (ATE); 12.760 -> 1.730 (RPE rot)Mixed
replace global alignment with C4D on static ScanNetDUSt3R · ScanNet (static) camera poseATE: 0.060 -> 0.064 (ATE); 0.751 -> 0.570 (RPE rot)Mixed
replace global alignment with C4D for video depthDUSt3R · Sintel video depthAbs Rel: 0.502 -> 0.478Improved
replace global alignment with C4D for video depthMASt3R · Sintel video depthAbs Rel: 0.370 -> 0.379Hurt
add camera movement alignment (CMA) objective (ablation)C4D · Sintel camera poseATE: 0.140 -> 0.103Improved
add camera trajectory smoothness (CTS) objective (ablation)C4D · Sintel camera poseRPE rot: 1.348 -> 0.705Improved
add point trajectory smoothness (PTS) objective (ablation)C4D · Sintel camera pose and depth metricspose and depth accuracy: ATE 0.103 -> 0.103; Abs Rel 0.327 -> 0.327No clear effect
combine CNN encoder with 3D-aware encoder in the point tracker (ablation)DynPT · TAP-Vid DAVISAJ: 42.4 -> 61.6 (vs 3D encoder only); 60.6 -> 61.6 (vs CNN only)Improved
use DynPT-guided motion masks in place of MonST3R's masksC4D · DAVIS2016 motion segmentationIoU: 31.57 -> 47.89Improved
predict mobility in addition to position and occlusion in the trackerDynPT vs CoTracker · TAP-Vid DAVISAJ: 61.8 -> 61.6No clear effect

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