AI Paper Atlas

Research map / 3D vision / Face and shape

SuperVoxelGPT: Adaptive and Ordered 3D Tokenization for Autoregressive Shape Generation

Yuan Li, Congyi Zhang, Xifeng Gao, Xiaohu Guo

ECCV 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 saliency-guided CVT instead of uniform CVT at same compression (ablation)SuperVoxelGPT · Image-to-3DCD_L2: 0.0157 -> 0.0122Improved
add saliency-guided CVT instead of uniform CVT (ablation)SuperVoxelGPT · Image-to-3DPSNR: 29.09 -> 32.08Improved
add SuperVoxelVAE local encoding instead of CraftsMan3D set-based encoder with FSQ (ablation)SuperVoxelGPT · Image-to-3DCD_L2: 0.0402 -> 0.0122Improved
add SuperVoxelVAE local encoding instead of set-based encoder (ablation)SuperVoxelGPT · Image-to-3DPSNR: 20.91 -> 32.08Improved
add SuperVoxelVAE encoding instead of set-based encoder (ablation)SuperVoxelGPT · Image-to-3DSpatial ordering compressibility gap: 0.0060 -> 0.1254Improved
add SuperVoxelVAE encoding instead of set-based encoder (ablation)SuperVoxelGPT · Image-to-3DAPCU: 0.5649 -> 0.7727Improved
use Jacobi decoding instead of token-by-token decoding (ablation)SuperVoxelGPT · Image-to-3DMLLM inference time (s): 47.21 -> 1.20Improved
use Jacobi decoding instead of token-by-token decoding (ablation)SuperVoxelGPT · Image-to-3DCD_L2 / PSNR: 0.0122 -> 0.0122; 32.09 -> 32.08No clear effect
use Jacobi decoding instead of token-by-token decoding (ablation)SuperVoxelGPT · Image-to-3DToken accuracy (%): 97.80 -> 97.50No clear effect
increase K and saliency threshold t (ablation)Saliency CVT · Parameter sweep (1,000 test shapes)compression vs reconstruction CD_L2: K=4, t=0.1 gives c=12.8%Mixed
use t=0.1 instead of t=0 (ablation)Saliency CVT · Parameter sweepreconstruction CD_L2No clear effect
use K=4 instead of K=6 or K=8 (ablation)Saliency CVT · Parameter sweepsalient-region fidelityImproved

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