DISN: Deep Implicit Surface Network for High-quality Single-view 3D Reconstruction
Authors: Qiangeng Xu, Weiyue Wang, Duygu Ceylan, Radomir Mech, Ulrich Neumann
NeurIPS 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We evaluate our approach on various shape categories using both synthetic data generated from 3D shape datasets as well as online product images. Qualitative and quantitative comparisons demonstrate that our network outperforms state-of-the-art methods and generates plausible shapes with high-quality details. We perform quantitative and qualitative comparisons on single-view 3D reconstruction with state-of-the-art methods [11 13, 16, 1] in Section 4.1. We also compare the performance of our method on camera pose estimation with [28] in Section 4.2. We further conduct ablation studies in Section 4.3 and showcase several applications in Section 4.4. |
| Researcher Affiliation | Collaboration | 1University of Southern California 2Adobe Los Angeles, California San Jose, California {weiyuewa,qiangenx,uneumann}@usc.edu {ceylan,rmech}@adobe.com |
| Pseudocode | No | No pseudocode or algorithm blocks are explicitly presented in the paper. |
| Open Source Code | Yes | Code is available at https://github.com/laughtervv/DISN. |
| Open Datasets | Yes | For both camera prediction and SDF prediction, we follow the settings of [11 13, 1], and use the Shape Net Core dataset [27], which includes 13 object categories, and an official training/testing split to train and test our method. We provide a new 2D dataset 1 composed of renderings of the models in Shape Net Core. 1https://github.com/Xharlie/Shapenet Render_more_variation |
| Dataset Splits | Yes | For both camera prediction and SDF prediction, we follow the settings of [11 13, 1], and use the Shape Net Core dataset [27], which includes 13 object categories, and an official training/testing split to train and test our method. |
| Hardware Specification | No | No specific hardware details (e.g., GPU/CPU models, memory) used for running experiments are provided in the paper. |
| Software Dependencies | No | Our network is implemented with Tensor Flow. We use VGG-16 [33] as the image encoder. Specific version numbers for software dependencies are not provided. |
| Experiment Setup | Yes | We choose m1 = 4, m2 = 1, and δ = 0.01 as the parameters of Equation 3. Our network is implemented with Tensor Flow. We use the Adam optimizer with a learning rate of 1 10 4 and a batch size of 16. |