Phidias: A Generative Model for Creating 3D Content from Text, Image, and 3D Conditions with Reference-Augmented Diffusion

Zhenwei Wang, Tengfei Wang, Zexin He, G. Hancke, Ziwei Liu +1 more
2/12/2026

No official code found. Best implementation: 3DTopia/Phidias-Diffusion (297 stars, MIT, last commit 2025-11-04). 1 more below.

Semantic Scholar

Code Implementations(2)

3DTopia/Phidias-DiffusionCommunity74%
MIT

[ICLR 2025] Phidias: A Generative Model for Creating 3D Content from Text, Image, and 3D Conditions with Reference-Augmented Diffusion

29715PythonSep 14, 202411 months agoMIT
3d-to-3ddiffusion-modelsimage-to-3dmulti-view-generationretrieval-augmented-generation+1 more
RAG-3D/RAG-3D.github.ioCommunity71%

Phidias: A Generative Model for Creating 3D Content from Text, Image, and 3D Conditions with Reference-Augmented Diffusion

00CSS, HTMLSep 13, 20241 years ago

Similar Papers

Learning with Partial-Label and Unlabeled Data: A Uniform Treatment for Supervision Redundancy and Insufficiency

95% match2026

UnionFormer: Unified-Learning Transformer with Multi-View Representation for Image Manipulation Detection and Localization

95% match2026

Propagate and Inject: Revisiting Propagation-Based Feature Imputation for Graphs with Partially Observed Features

94% match1 repo2026

GTR: A General, Multi-View, and Dynamic Framework for Trajectory Representation Learning

94% match2026

WildlifeMapper: Aerial Image Analysis for Multi-Species Detection and Identification

94% match1 repo332026

Hypervolume Knowledge Gradient: A Lookahead Approach for Multi-Objective Bayesian Optimization with Partial Information

93% match2026

Cite this paper

@article{wang2026phidias,
  title  = {Phidias: A Generative Model for Creating 3D Content from Text, Image, and 3D Conditions with Reference-Augmented Diffusion},
  author = {Zhenwei Wang and Tengfei Wang and Zexin He and G. Hancke and Ziwei Liu and Rynson W. H. Lau},
  year   = {2026},
  url    = {https://api.semanticscholar.org/CorpusID:69e5eec04d7f988c5ef1d325849460eb9ca2d350},
  journal = {ICLR 2025 2025}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/69e5eec04d7f988c5ef1d325849460eb9ca2d350.svg)](https://codeofpaper.com/paper/69e5eec04d7f988c5ef1d325849460eb9ca2d350)

Discussion