Learning Implicit Body Representations from Double Diffusion Based Neural Radiance Fields
Authors: Guangming Yao, Hongzhi Wu, Yi Yuan, Lincheng Li, Kun Zhou, Xin Yu
IJCAI 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Experiments on various datasets demonstrate that our approach outperforms the state-of-the-art in both geometric reconstruction and novel view synthesis. We perform experiments on the synthesized datasets Twindom, THuman2.0 [Zheng et al., 2021] and the real-world dataset ZJU-Mocap [Peng et al., 2021]. |
| Researcher Affiliation | Collaboration | Guangming Yao1 , Hongzhi Wu1 , Yi Yuan2 , Lincheng Li2 , Kun Zhou1 and Xin Yu3 1Zhejiang University 2Net Ease Fuxi AI Lab 3University of Technology Sydney |
| Pseudocode | No | The paper does not contain any structured pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not provide an explicit statement or link for open-source code for the described methodology. |
| Open Datasets | Yes | We perform experiments on the synthesized datasets Twindom1, THuman2.0 [Zheng et al., 2021] and the real-world dataset ZJU-Mocap [Peng et al., 2021]. ... 1web.twindom.com |
| Dataset Splits | Yes | We use 1,200 body meshes from Twindom for training, 300 meshes for evaluation. For THuman, 400 meshes are used for training, and 100 meshes for evaluation. |
| Hardware Specification | Yes | while Neural Body requires 4 hours to learn from scratch for each subject on an Nvidia RTX3090 GPU. |
| Software Dependencies | No | The paper mentions using the Adam optimizer but does not provide specific version numbers for software dependencies or libraries. |
| Experiment Setup | Yes | We use the Adam optimizer [Kingma and Ba, 2014], and set the learning rate to 1 × 10−4. The loss weights are set to 10, 1 for λr and λe, respectively. |