Leveraging Pretrained Diffusion Models for Zero-Shot Part Assembly
Authors: Ruiyuan Zhang, Qi Wang, Jiaxiang Liu, Yuchi Huo, Chao Wu
IJCAI 2025 | Venue PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | To verify our work, we conduct extensive experiments and quantitative comparisons to several strong baseline methods, demonstrating the effectiveness of the proposed approach, which even surpasses the supervised learning method. [...] 5 Experiments |
| Researcher Affiliation | Academia | 1Zhejiang University 2North China Electric Power University EMAIL, EMAIL |
| Pseudocode | No | The paper describes the methodology in Section 3 (Methodology), using mathematical equations and textual descriptions, but does not include any structured pseudocode or algorithm blocks. |
| Open Source Code | Yes | The code has been released on https:// github.com/Ruiyuan-Zhang/Zero-Shot-Assembly. |
| Open Datasets | Yes | We conduct experiments on the Chair subset of Part Net [2019], a large-scale dataset with fine-grained part-level annotations. [...] [Mo et al., 2019] Kaichun Mo, Shilin Zhu, Angel X Chang, Li Yi, Subarna Tripathi, Leonidas J Guibas, and Hao Su. Partnet: A large-scale benchmark for fine-grained and hierarchical part-level 3d object understanding. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 909 918, 2019. |
| Dataset Splits | Yes | We follow the official train/val/test splits: the training set is used to train a diffusion model for 3D point cloud generation, and the test set is used for zero-shot assembly. |
| Hardware Specification | No | The paper mentions running experiments and training models but does not provide any specific details about the hardware used (e.g., GPU models, CPU types, memory). |
| Software Dependencies | No | The paper mentions using "stable diffusion 2.1 [Rombach et al., 2022]" for certain experiments, which is a specific model version. However, it does not provide a reproducible description of the ancillary software environment, such as programming languages or libraries with version numbers (e.g., Python, PyTorch/TensorFlow versions). |
| Experiment Setup | Yes | Therefore, we fixed the time step z to a small value, typically 2 or 4, to obtain more accurate ICP estimates. [...] We evaluated our method against various baselines, as shown in Table 1, Fig. 2, and the Appendix. Our approach outperforms current methods in addressing the zeroshot challenge. All of our metrics outperform existing methods, except for SCD. This is expected since our method emphasizes density estimates from a diffusion model rather than sampling complete shapes. [...] We utilized stable diffusion 2.1 [Rombach et al., 2022] with the prompt a picture of colorful chair to generate a realistic chair shape. As shown in Fig. 8, we demonstrate the visualization of the assembly process. These images are processed through microrendering, serving as inputs for the diffusion model to obtain SDS loss. |