AI Paper Atlas

Research map / 3D vision / Face and shape

Pixel2Mesh++: 3D Mesh Generation and Refinement From Multi-View Images

Chao Wen, Yinda Zhang, Chenjie Cao, Zhuwen Li, Xiangyang Xue, Yanwei Fu

TPAMI 2022 · 3D vision › Face and shape · approach: 3D shape generative modeling · 48 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 Multi-View Deformation Network refinement on top of MVP2M coarse meshMVP2M · ShapeNet multi-view (mean over 13 categories)F-score(tau): 61.20 -> 67.23Improved
Add MDN refinement on top of MVDISN coarse meshMVDISN · ShapeNet multi-view (mean over 13 categories)F-score(tau): 70.78 -> 75.24Improved
Add MDN refinement on top of MVP2MMVP2M · ShapeNet multi-view (mean over 13 categories)Chamfer Distance: 0.456 -> 0.381Improved
Add MDN refinement on top of MVDISNMVDISN · ShapeNet multi-view (mean over 13 categories)volumetric IoU: 0.527 -> 0.508Hurt
Add MDN refinement on top of MVP2MMVP2M · ShapeNet multi-view (mean over 13 categories)volumetric IoU: 0.411 -> 0.436Improved
Add multi-view feature pooling to Pixel2Mesh (MVP2M vs P2M-M)P2M-M · ShapeNet multi-view (mean over 13 categories)F-score(tau): 61.72 -> 61.20No clear effect
Add feature statistics (mean, max, std) instead of concatenation (ablation)Ours-P · ShapeNet ablation (Ours-P, known camera)F-score(tau): 65.26 -> 67.32Improved
Add feature statistics instead of concatenation (ablation)Ours-P · ShapeNet ablation (Ours-P, known camera)Chamfer Distance: 0.511 -> 0.381Improved
Add re-sampled Chamfer loss instead of vertex-only Chamfer (ablation)Ours-P · ShapeNet ablation (Ours-P, known camera)F-score(tau): 66.26 -> 67.32Improved
Add re-sampled Chamfer loss (ablation)Ours-P · ShapeNet ablation (Ours-P, known camera)Chamfer Distance: 0.496 -> 0.381Improved
Increase test views from 2 to 5 (trained on 3 views)Ours-P (MDN) · ShapeNet viewsF-score(tau): 64.48 -> 68.29Improved
Use predicted camera pose instead of ground-truthOurs-P · ShapeNet robustnessF-score(tau): 67.32 -> 63.22Hurt

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