Research map / 3D vision / 3D generation
CoGrad3D: Spatially-Coupled Timestep Optimization with Orthogonal Gradient Fusion for 3D Generation
Haoyang Tong, Hongbo Wang, Jin Liu, Qi Wang, Jie Cao, Ran He
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add adaptive region sampling (ablation) | CoGrad3D · DreamFusion gallery (ablation) | CLIP-S B/32: 28.28 -> 31.42 | Improved |
| add adaptive region sampling (ablation) | CoGrad3D · DreamFusion gallery (ablation) | ImageReward: -0.0447 -> 0.7449 | Improved |
| add hybrid adaptive timestep sampling (ablation) | CoGrad3D · DreamFusion gallery (ablation) | CLIP-S B/32: 29.03 -> 31.42 | Improved |
| add hybrid adaptive timestep sampling (ablation) | CoGrad3D · DreamFusion gallery (ablation) | ImageReward: -0.4829 -> 0.7449 | Improved |
| add orthogonal gradient fusion (ablation) | CoGrad3D · DreamFusion gallery (ablation) | CLIP-S B/32: 27.97 -> 31.42 | Improved |
| add orthogonal gradient fusion (ablation) | CoGrad3D · DreamFusion gallery (ablation) | VQAscore: 0.5541 -> 0.6286 | Improved |
| add orthogonal gradient fusion (ablation) | CoGrad3D · DreamFusion gallery (ablation) | CLIP-S L/14: 25.21 -> 28.91 | Improved |
| CoGrad3D versus LucidDreamer on Janus rate | LucidDreamer · DreamFusion gallery (80 prompts) | Janus rate (JR): 0.2000 -> 0.2250 | Hurt |
| CoGrad3D versus best baseline on text-3D alignment | LucidDreamer · DreamFusion gallery (80 prompts) | CLIP-S B/32: 30.02 -> 31.42 | Improved |
| CoGrad3D versus best baseline on user rank | LucidDreamer · DreamFusion gallery (80 prompts) | User rank: 3.18 -> 2.16 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- GaussianDreamer — GaussianDreamer: Fast Generation from Text to 3D Gaussians by Bridging 2D and 3D Diffusion Models (CVPR 2024) · better on 6 of 6 shared results
- LucidDreamer — LucidDreamer: Towards High-Fidelity Text-to-3D Generation via Interval Score Matching (CVPR 2024) · better on 5 of 6 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
- DreamTime: An Improved Optimization Strategy for Diffusion-Guided 3D Generation (ICLR 2024 · similarity 0.89)
- Consistent3D: Towards Consistent High-Fidelity Text-to-3D Generation with Deterministic Sampling Prior (CVPR 2024 · similarity 0.89)
- HIFA: High-fidelity Text-to-3D Generation with Advanced Diffusion Guidance (ICLR 2024 · similarity 0.89)
- Instant3D: Fast Text-to-3D with Sparse-view Generation and Large Reconstruction Model (ICLR 2024 · similarity 0.88)
- ViewDiff: 3D-Consistent Image Generation with Text-to-Image Models (CVPR 2024 · similarity 0.88)
- Let 2D Diffusion Model Know 3D-Consistency for Robust Text-to-3D Generation (ICLR 2024 · similarity 0.88)
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.
