AI Paper Atlas

Research map / 3D vision / 3D generation

DiffuStereo: High Quality Human Reconstruction via Diffusion-Based Stereo Using Sparse Cameras

Ruizhi Shao, Zerong Zheng, Hongwen Zhang, Jingxiang Sun, Yebin Liu

ECCV 2022 · 3D vision › 3D generation · approach: Diffusion-based novel view synthesis · 57 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
remove diffusion model from the stereo network (add diffusion) (ablation)DiffuStereo stereo network · THuman2.0 stereo (45 deg)average disparity error (pixels): 1.448 -> 0.632Improved
use the proposed diffusion kernel instead of the original 1000-step kernel (ablation)DiffuStereo · THuman2.0 stereo (45 deg)average disparity error (pixels): 0.881 -> 0.632Improved
use the proposed diffusion kernel instead of the original kernel (ablation)DiffuStereo · THuman2.0 stereo (45 deg)disparity error within 1/2 pixel (%): 52.9 -> 67.6Improved
add diffusion-based stereo refinement on top of RAFT-Stereo given the same coarse depth DcRAFT-Stereo (w. Dc) · THuman2.0 stereo (45 deg)average disparity error (pixels): 1.567 -> 0.632Improved
use the proposed diffusion kernel (original kernel unstable) (ablation)DiffuStereo · depth mapsdepth map stabilityImproved
reduce reverse diffusion to about 10 steps with the new kernel (ablation)DiffuStereo · depth mapsdetail recoveryNo clear effect
use the proposed 30-step diffusion instead of the original 1000-step model (ablation)DiffuStereo · inference timetime: 12s with 30 diffusion stepsImproved
use visual hull from 8 views as the initial coarse model instead of DoubleField mesh (ablation)DiffuStereo · coarse model initializationvisible-region detailNo clear effect
increase view angle between stereo pair from 20 to 45 degDiffuStereo · THuman2.0 stereodisparity error within 1/2 pixel (%): 71.3 -> 67.6Hurt
replace DoubleField reconstruction with DiffuStereo refinementDoubleField · THuman2.0 reconstructionChamfer distance (mm): 2.879 -> 1.198Improved
replace DoubleField reconstruction with DiffuStereo refinementDoubleField · THuman2.0 reconstructionratio within 1mm (%): 23.2 -> 68.1Improved

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