AI Paper Atlas

Research map / 3D vision / Face and shape

MeshWeaver: Sparse-Voxel-Guided Surface Weaving for Autoregressive Mesh Generation

Jiale Xu, Wang Zhao, Ying Shan

CVPR 2026 · 3D vision › Face and shape · approach: Textured mesh generation · 1 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 voxel features as vertex representation (ablation)MeshWeaver · Toys4KCD (x10^-1): 0.142 -> 0.116Improved
add cross-attention-guided token prediction (ablation)MeshWeaver · Toys4KCD (x10^-1): 0.146 -> 0.116Improved
add both voxel features and cross-attention (ablation)MeshWeaver · Toys4KCD (x10^-1): 0.158 -> 0.116Improved
add sparse-voxel scaffold masking at inference (ablation)MeshWeaver · Toys4KCD (x10^-1): 0.122 -> 0.116Improved
add sparse-voxel scaffold masking at inference (ablation)MeshWeaver · Toys4KHD: 0.090 -> 0.087Improved
add voxel features as vertex representation (ablation)MeshWeaver · Toys4KNC: 0.694 -> 0.732Improved
use two-level (16,8) instead of three-level (8,4,4) space partition (ablation)MeshWeaver · Toys4KCD (x10^-1): 0.137 -> 0.116Improved
use (16,8) instead of (8,16) space partition (ablation)MeshWeaver · Toys4KCD (x10^-1): 0.120 -> 0.116No clear effect
allocate 20 instead of 16 layers to level 0 (ablation)MeshWeaver · Toys4KCD (x10^-1): 0.116 -> 0.121No clear effect
allocate 18 instead of 16 layers to level 0 (ablation)MeshWeaver · Toys4KCD (x10^-1): 0.116 -> 0.113No clear effect
add cross-attention KV cache (ablation)MeshWeaver · Toys4K (200 meshes)throughput (tokens/s): 26.8 -> 30.7Improved
use vertex-level tokenization instead of coordinate-levelMeshWeaver · mesh tokenizationcompression ratio: 0.22 -> 0.18Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 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