Research map / Image generation / Text-to-image diffusion
LEGO-Maker: A Semantic-Driven Algorithm for Text-to-3D Generation
Yifei Zhang, Lei Chen
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 texture enhancement (ablation) | LEGO-Maker (TRELLIS) · T3Bench (Multiple Objects), ablation | Quality: 51.25 -> 52.90 | Improved |
| add texture enhancement (ablation) | LEGO-Maker (TRELLIS) · T3Bench (Multiple Objects), ablation | Consistency: 42.89 -> 43.60 | Improved |
| add texture enhancement (ablation) | LEGO-Maker (SPAR3D) · T3Bench (Multiple Objects), ablation | Quality: 49.47 -> 51.17 | Improved |
| add diffusion retraining (ablation) | LEGO-Maker (TRELLIS) · T3Bench (Multiple Objects), ablation | Quality: 49.04 -> 52.90 | Improved |
| add diffusion retraining (ablation) | LEGO-Maker (TRELLIS) · T3Bench (Multiple Objects), ablation | Consistency: 39.58 -> 43.60 | Improved |
| add diffusion retraining (ablation) | LEGO-Maker (SPAR3D) · T3Bench (Multiple Objects), ablation | Quality: 48.26 -> 51.17 | Improved |
| add both texture enhancement and diffusion retraining (ablation) | LEGO-Maker (TRELLIS) · T3Bench (Multiple Objects), ablation | Quality: 48.23 -> 52.90 | Improved |
| add both texture enhancement and diffusion retraining (ablation) | LEGO-Maker (SPAR3D) · T3Bench (Multiple Objects), ablation | Consistency: 36.89 -> 41.87 | Improved |
| use TRELLIS instead of SPAR3D as image-to-3D generator (ablation) | LEGO-Maker · T3Bench (Multiple Objects), ablation | Quality: 51.17 -> 52.90 | No clear effect |
| use LEGO-Maker instead of Meshy 2.0 | Meshy 2.0 · T3Bench (Multiple Objects) | Quality: 49.33 -> 52.90 | Improved |
| use LEGO-Maker instead of Hun Yuan 2.0 | Hun Yuan 2.0 · Toys4K | FD_dinov2: 249.75 -> 210.23 | Improved |
| use LEGO-Maker instead of Hun Yuan 2.0 for speed | Hun Yuan 2.0 · Toys4K | Inference time: 1 min -> 3 min | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- DreamFusion — DreamFusion: Text-to-3D using 2D Diffusion (ICLR 2023) · better on 4 of 4 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
- Progressive3D: Progressively Local Editing for Text-to-3D Content Creation with Complex Semantic Prompts (ICLR 2024 · similarity 0.89)
- Let Language Constrain Geometry: Vision–Language Models as Semantic and Spatial Critics for 3D Generation (ICML 2026 · similarity 0.88)
- 3D-SceneDreamer: Text-Driven 3D-Consistent Scene Generation (CVPR 2024 · similarity 0.88)
- ViewDiff: 3D-Consistent Image Generation with Text-to-Image Models (CVPR 2024 · similarity 0.88)
- EfficientDreamer: High-Fidelity and Robust 3D Creation via Orthogonal-view Diffusion Priors (CVPR 2024 · similarity 0.88)
- Compositional 3D-aware Video Generation with LLM Director (NeurIPS 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.
