AI Paper Atlas

Research map / 3D vision / 3D generation

HiFi-123: Towards High-fidelity One Image to 3D Content Generation

Wangbo Yu, Li Yuan, Yan-Pei Cao, Xiangjun Gao, Xiaoyu Li, Wenbo Hu, Long Quan, Ying Shan, Yonghong Tian

ECCV 2024 · 3D vision › 3D generation · approach: Score distillation for text-to-3D · 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
apply RGNV enhancement to Zero-1-to-3 novel viewsZero-1-to-3 · GSO novel view synthesisPSNR: 18.95 -> 20.45Improved
apply RGNV enhancement to SyncDreamer novel viewsSyncDreamer · single view datasetCLIP similarity: 0.851 -> 0.896Improved
use RGSD loss instead of SDS loss in the refine stage (ablation)HiFi-123 · GSO 3D datasetPSNR: 19.25 -> 23.68Improved
use RGSD loss instead of SDS loss in the refine stage (ablation)HiFi-123 · Single view datasetContextual distance: 1.925 -> 1.627Improved
refine stage with SDS loss over coarse stage (ablation)HiFi-123 · Single view datasetContextual distance: 1.901 -> 1.925No clear effect
use depth-conditioned DDIM inversion rather than regular DDIM inversion (ablation)RGNV · 400 images reconstructionLPIPS: 0.2835 -> 0.0661Improved
remove attention injection from RGNV or use SDEdit instead (ablation)RGNV · qualitative ablationvisual consistencyHurt
use RGSD loss with alternating fixed-view targets for faster refinementHiFi-123 refine stage · training timeminutes: 60 -> 10Improved
naive image loss using RGNV-enhanced rendered views (ablation)HiFi-123 refine stage · qualitativetexture sharpnessHurt
increase attention injection start step l (ablation)RGNV · qualitativetexture fidelity vs geometryMixed

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 4 of 4 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