AI Paper Atlas

Research map / 3D vision / 3D generation

VQ3D: Learning a 3D-Aware Generative Model on ImageNet

Kyle Sargent, Jing Yu Koh, Han Zhang, Huiwen Chang, Charles Herrmann, Pratul Srinivasan, Jiajun Wu, Deqing Sun

ICCV 2023 · 3D vision › 3D generation · approach: Feed-forward 3D asset generation and editing · 26 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
replace GAN-based 3D generator with vector-quantized autoencoder plus NeRF decoder3D-aware generative model · ImageNet generationFID: 69.8 -> 16.8Improved
add pseudo-GT depth loss to baseline GAN methods (ablation)pi-GAN, GIRAFFE, EG3D · ImageNet generationFID: EG3D 109.3 -> 91.8; GIRAFFE 132.1 -> 132.0No clear effect
add weighted pointwise depth loss (ablation)VQ3D Stage 2 · ImageNet generationDepth accuracy: 1.90 -> 0.16Improved
remove depth loss from Stage 1 (ablation)VQ3D Stage 1 · ImageNet reconstructionFID / depth accuracy: FID 11.2 -> 4.0; depth acc. 0.18 -> 1.91Mixed
use triplanes instead of a fully implicit representation (ablation)VQ3D Stage 1 · ImageNet reconstructionFID: 273.5 -> 11.2Improved
use ViT encoder and decoder instead of CNN (ablation)VQ3D Stage 1 · ImageNet reconstructiontraining stabilityImproved
add GAN loss and depth scale loss (ablation)VQ3D Stage 1 · ImageNet reconstructiondisparity scale / novel view quality: disparity scale 1.27 -> 1.00 (GAN); 1.21 -> 1.00 (scale)Improved
add NeRF distortion and interlevel losses (ablation)VQ3D Stage 1 · ImageNet reconstructiondepth accuracy: 0.28 -> 0.18Improved
restrict sampling to top-k 1000 instead of the full 8192 codebook (ablation)VQ3D Stage 2 · ImageNet generationFID: 36.1 -> 31.5Improved
use overly restrictive top-p (0.95) (ablation)VQ3D Stage 2 · ImageNet generationFID: 32.2 (p=0.98) -> 35.7 (p=0.95)Hurt
VQ3D vs best baselines on CompCars single-class datasetVQ3D · CompCarsFID: 7.2 -> 7.3 (vs GIRAFFE HD)No clear effect

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