AI Paper Atlas

Research map / 3D vision / Face and shape

Learning Versatile 3D Shape Generation with Improved Auto-regressive Models

Simian Luo, Xuelin Qian, Yanwei Fu, Yinda Zhang, Ying Tai, Zhenyu Zhang, Chengjie Wang, Xiangyang Xue

ICCV 2023 · 3D vision › Face and shape · approach: 3D shape generative modeling · 5 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 coupling network to project three planes into a latent vector before quantization (ablation)ImAM · ShapeNet plane1-NNA: 73.67 -> 59.95Improved
add coupling network to project three planes into a latent vector before quantization (ablation)ImAM · ShapeNet planeECD: 743 -> 236Improved
quantize latent vector instead of volumetric grid (ablation)ImAM · ShapeNet reconstructionIoU: 88.87 -> 88.01No clear effect
quantize latent vector instead of volumetric grid (transformer trainability) (ablation)ImAM · ShapeNet generation stage 2trainabilityImproved
increase feature resolution from 16 to 32 (ablation)ImAM stage 1 · ShapeNet reconstructionIoU: 79.17 -> 88.01Improved
increase codebook entries from 1024 to 4096 (ablation)ImAM stage 1 · ShapeNet reconstructionIoU: 86.57 -> 88.01Improved
use ImAM instead of GBIF for unconditional generationImAM · ShapeNet unconditional (5 categories, avg)ECD: 858 -> 240Improved
use ImAM vs PVD on Car categoryImAM · ShapeNet unconditional (Car)1-NNA: 57.37 (PVD) vs 76.58Hurt
use ImAM instead of GBIF on COVImAM · ShapeNet unconditional (5 categories, avg)COV: 80.57 -> 78.29Hurt
use ImAM instead of AutoSDF on bottom-half completion MMDImAM · Partial point completion, bottom-half (avg)MMD (x10^3): 1.751 -> 1.837Hurt
use ImAM instead of CLIP-Forge for prompt-based text-to-shape FPDImAM · Text-guided generation, prompts (Text2Shape)FPD: 4.14 -> 5.39Hurt

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