AI Paper Atlas

Research map / 3D vision / Depth and stereo

MotionCrafter: Dense Geometry and Motion Reconstruction with a 4D VAE

Ruijie Zhu, Jiahao Lu, Wenbo Hu, Xiaoguang Han, Jianfei Cai, Ying Shan, Chuanxia Zheng

CVPR 2026 · 3D vision › Depth and stereo · approach: Learned optical and scene flow · 5 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
fine-tune the whole VAE with mean normalization instead of decoder-only with max normalization (ablation)Geometry VAE · SintelRel^p: 20.76 -> 5.68Improved
fine-tune the whole VAE with mean normalization instead of decoder-only with max normalization (ablation)Geometry VAE · MonkaaRel^p: 14.44 -> 5.03Improved
use mean-rescaled VAE in the U-Net stage instead of max-rescaled (ablation)Geometry U-Net · SintelRel^p: 40.47 -> 35.17Improved
use mean-rescaled VAE in the U-Net stage instead of max-rescaled (ablation)Geometry U-Net · MonkaaRel^p: 33.66 -> 27.36Improved
add depth-map supervision to Geometry VAE training (ablation)Geometry VAE · ScanNetRel^p: 4.46 -> 3.03Improved
add depth-map supervision to Geometry VAE training (ablation)Geometry VAE · SintelRel^p: 5.68 -> 4.39Improved
add decoder loss when training the U-Net (ablation)Geometry U-Net · DDADRel^p: 33.78 -> 22.39Improved
add decoder loss when training the U-Net (ablation)Geometry U-Net · Sinteldelta^p: 58.08 -> 56.94Hurt
use unified fusion of geometry and motion latents instead of separate VAEs (ablation)Motion U-Net · SpringEPE: 6.37 -> 5.16Improved
use unified fusion of geometry and motion latents instead of separate VAEs (ablation)Motion U-Net · Point OdysseyEPE: 3.65 -> 3.49Improved
use unified fusion instead of separate motion VAE (ablation)Motion VAE · SpringEPE: 0.66 -> 0.88Hurt
use deterministic paradigm instead of denoising diffusion paradigm (ablation)MotionCrafter · SintelRel^p: 35.95 -> 32.46Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 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