AI Paper Atlas

Research map / 3D vision / Face and shape

Learning Signed Distance Field for Multi-View Surface Reconstruction

Jingyang Zhang, Yao Yao, Long Quan

ICCV 2021 · 3D vision › Face and shape · approach: Neural implicit surface reconstruction · 117 citations (OpenAlex)

Read the paper

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 feature loss L_F (ablation)MVSDF (distance only) · DTUChamfer distance (mm): 3.56 -> 1.75Improved
add render loss L_R (ablation)MVSDF (distance only) · DTUChamfer distance (mm): 3.56 -> 1.06Improved
add render loss on top of feature loss (ablation)MVSDF · DTUChamfer distance (mm): 1.75 -> 0.88Improved
add feature loss on top of render loss (ablation)MVSDF · DTUChamfer distance (mm): 1.06 -> 0.88Improved
use lower-resolution MVS depth maps (ablation)MVSDF · DTUChamfer distance (mm): 0.88 -> 0.89No clear effect
use precomputed geometrically filtered depth maps (ablation)MVSDF · DTUChamfer distance (mm): 0.88 -> 0.88No clear effect
use MVSDF instead of IDR without masksIDR · DTUPSNR: 23.20 -> 25.92Improved
use MVSDF instead of Vis-MVSNetVis-MVSNet · DTUChamfer distance (mm): 0.88 -> 0.88No clear effect
use MVSDF instead of IDRIDR · EPFLChamfer distance (x10^-2): 25.30 -> 6.61Improved
use MVSDF instead of Vis-MVSNetVis-MVSNet · Tanks and TemplesPSNR: 19.56 -> 24.54Improved

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