AI Paper Atlas

Research map / 3D vision / 3D generation

VFusion3D: Learning Scalable 3D Generative Models from Video Diffusion Models

Junlin Han, Filippos Kokkinos, Philip Torr

ECCV 2024 · 3D vision › 3D generation · approach: Score distillation for text-to-3D · 49 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
use VFusion3D trained on synthetic multi-view data instead of OpenLRM or LGMVFusion3D · GSO (996 objects)PSNR: 14.21 -> 20.89Improved
increase 3D fine-tuning data for Emu Video from 10K to 100K (ablation)Emu Video fine-tuning · good rate of generated multi-view videosgood rate: 44.5% -> 61.3%Improved
add multi-stage training recipe (ablation)VFusion3D · 500K synthetic dataLPIPS: 0.206 -> 0.168Improved
remove pixel-level (MSE) image supervision (ablation)VFusion3D · 500K synthetic dataCLIP text similarity: 0.249 -> 0.257Improved
add opacity supervision (ablation)VFusion3D · 500K synthetic dataCLIP image similarity: 0.798 -> 0.802Improved
add camera noise injection (ablation)VFusion3D · 500K synthetic dataCLIP text similarity: 0.256 -> 0.252No clear effect
fine-tune VFusion3D with more 3D data (none to 100K) (ablation)VFusion3D · test 3D modelsLPIPS: 0.160 -> 0.143Improved
select 3D fine-tuning data by aesthetic score instead of random (ablation)VFusion3D · test 3D modelsLPIPS: 0.147 vs 0.146 (random)No clear effect
use 32 rendered views instead of 16 for 3D fine-tuning (ablation)VFusion3D · test 3D modelsSSIM: 0.838 -> 0.839No clear effect
train on synthetic multi-view data instead of 150K real 3D data (ablation)VFusion3D · text-to-3D generalizationCLIP image similarity: 0.631 -> 0.839Improved
combine synthetic pre-training with 3D data fine-tuning (ablation)VFusion3D · test 3D modelsLPIPS: 0.160 -> 0.143Improved
increase synthetic dataset size from 100K to 2.7M (ablation)VFusion3D · scaling studyLPIPS and CLIP image similarityImproved

Later papers that compared with this method

2 later accepted papers report VFusion3D as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.

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