AI Paper Atlas

Research map / 3D vision / Face and shape

Differentiable Convex Polyhedra Optimization from Multi-view Images

Daxuan Ren, Haiyi Mei, Hezi Shi, Jianmin Zheng, Jianfei Cai, Lei Yang

ECCV 2024 · 3D vision › Face and shape · approach: Textured mesh generation · 1 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 convex densification via Loop subdivision (ablation)DiffConvex with 16 convexes · ablation shapeCD L2: 0.46 -> 0.32Improved
add random convex spawning (ablation)DiffConvex with 16 convexes · ablation shapeCD L2: 0.54 -> 0.32Improved
increase number of convexes from 16 to 128 (ablation)DiffConvex · ablation shapeCD L2: 0.32 -> 0.05Improved
increase number of convexes from 128 to 512 (ablation)DiffConvex · ablation shapeCD L2: 0.05 -> 0.05No clear effect
image-supervised differentiable convex polyhedra instead of implicit-field-supervised CVXNet/BSPNetShapeNet reconstruction · ShapeNet meanL1 Chamfer Distance: 0.024 (CVXNet) -> 0.022Improved
image-supervised convexes instead of CVXNetShapeNet reconstruction · ShapeNet meanL2 Chamfer Distance x1000: 0.526 (CVXNet) -> 0.592Hurt
image-only supervision on categories with large cavitiesDiffConvex · ShapeNet cabinetL1 Chamfer Distance: 0.026 (CVXNet) -> 0.042Hurt
convex polyhedra with volumetric texture instead of superquadric blocks (DBW)multiview reconstruction · NeRF Synthetic and DTUL2 Chamfer Distance: Lego 0.91 -> 0.44; DTU S105 5.21 -> 2.62Improved
convex polyhedra instead of DBWmultiview reconstruction · DTU S24L2 Chamfer Distance: 3.25 -> 3.87Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 2 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