AI Paper Atlas

Research map / 3D vision / Depth and stereo

3DFG-PIFu: 3D Feature Grids for Human Digitization from Sparse Views

Kennard Yanting Chan, Fayao Liu, Guosheng Lin, Chuan Sheng Foo, Weisi Lin

ECCV 2024 · 3D vision › Depth and stereo · approach: Cost-volume multi-view stereo · 0 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
add 3D visual hull feature grid GM to single-view PIFu (ablation)PIFu · THuman2.0Chamfer distance (10^-5): 26.97 -> 6.626Improved
add front-normal grid GN alongside GM (ablation)PIFu + GM · THuman2.0Chamfer distance (10^-5): 6.626 -> 6.007Improved
add view visibility grid GV to 2nd stage (ablation)3DFG-PIFu 2nd stage · THuman2.0 visible regionsChamfer distance (10^-5): 4.036 -> 3.891Improved
add iterative 2nd stage refinement (ablation)3DFG-PIFu · THuman2.0Chamfer distance (10^-5): 6.007 -> 5.796Improved
add iterative 2nd stage refinement (ablation)3DFG-PIFu · BUFFChamfer distance (10^2): 2.386 -> 2.509Hurt
use SDF-based SMPL-X features GX instead of S-PIFu features (ablation)single-view PIFu · THuman2.0Chamfer distance (10^-5): 4.488 -> 4.352Improved
combine GX with PaMIR voxel-aligned features (ablation)single-view PIFu · THuman2.0Chamfer distance (10^-5): 4.104 -> 3.970Improved
combine GX with voxel-aligned features (ablation)single-view PIFu · BUFFChamfer distance (10^2): 8.037 -> 8.012Mixed
use ground-truth SMPL-X mesh and high-res images in 3DFG-PIFu3DFG-PIFu · BUFFChamfer distance (10^2): 2.509 -> 3.412Mixed
replace point-wise multi-view fusion with 3D feature grids3DFG-PIFu vs SeSDF · THuman2.0Chamfer distance (10^-5): 6.302 -> 5.796Improved

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