AI Paper Atlas

Research map / 3D vision / 3D generation

View Selection for 3D Captioning via Diffusion Ranking

Tiange Luo, Justin Johnson, Honglak Lee

ECCV 2024 · 3D vision › 3D generation · approach: Text-to-3D scene generation · 30 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
select top-6 views with DiffuRank instead of the bottom-6 views for GPT4-Vision captioning (ablation)DiffuRank + GPT4-Vision captioning · Objaverse captions, quality A/B testquality score (1-5): 2.74 -> lower is better for the other caption; bottom-6 views lose to ours with 31.1 win / 52.0 loseImproved
use DiffuRank top-6 views instead of all 28 views (ablation)GPT4-Vision captioning · Objaverse captions, CLIP scoreCLIP score: 73.5 -> 74.6Improved
use DiffuRank top-6 views instead of horizontal 6 views (ablation)GPT4-Vision captioning · Objaverse captions, CLIP R@1CLIP R@1: 25.8 -> 26.7Improved
use DiffuRank top-6 views instead of bottom-6 views (ablation)GPT4-Vision captioning · Objaverse captions, hallucination A/B testhallucination score (1-5): 2.61 vs 2.85 (all views) and 2.90 (horizontal); bottom-6 loses 57.0%Improved
replace Cap3D captions with DiffuRank + GPT4-Vision captionsObjaverse captions · Objaverse captions (5k objects)CLIP R@1: 20.5 -> 26.7Improved
fine-tune Shap-E (STF) on Cap3D captionsShap-E (STF) · Text-to-3D finetuningCLIP R@1: 12.7 -> 11.9Hurt
fine-tune Shap-E (NeRF) on Cap3D captionsShap-E (NeRF) · Text-to-3D finetuningCLIP R@1: 12.2 -> 11.7Hurt
use corrected captions (330k, ~20% updated) instead of Cap3D captionsShap-E (STF) · Text-to-3D finetuningCLIP R@1: 11.9 -> 13.4Improved
scale training captions from 330k to 825k (ablation)Shap-E (STF) · Text-to-3D finetuningCLIP R@1: 13.4 -> 14.9Improved
scale training captions from 330k to 825k (ablation)Point-E · Text-to-3D finetuningCLIP R@1: 8.1 -> 10.1Improved
fine-tune Point-E on Cap3D captionsPoint-E · Text-to-3D finetuningCLIP R@1: 3.4 -> 7.1Improved

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