SC-NeuS: Consistent Neural Surface Reconstruction from Sparse and Noisy Views
Authors: Shi-Sheng Huang, Zixin Zou, Yichi Zhang, Yan-Pei Cao, Ying Shan
AAAI 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | With extensive evaluation on public datasets, our SCNeu S can achieve consistently better surface reconstruction results with fine-grained details than previous approaches, especially from sparse and noisy camera views. |
| Researcher Affiliation | Collaboration | 1Beijing Normal University 2Tsinghua University 3Beijing Institute of Technology 4 ARC Lab, Tencent PCG |
| Pseudocode | No | No structured pseudocode or algorithm blocks were found. |
| Open Source Code | Yes | The source code is available at https://github.com/zouzx/sc-neus.git. |
| Open Datasets | Yes | To evaluate the effectiveness of our SC-Neu S, we conduct extensive experiments on public dataset including DTU (Jensen et al. 2014) and Blended MVS (Yao et al. 2020) with various geometry scenarios. |
| Dataset Splits | No | No explicit train/validation/test dataset splits with percentages or counts are provided for reproducibility, beyond stating 'we choose to evaluate our approach on the public DTU dataset (Aanæs et al. 2016) with 15 different object scan. For sparse views, we follow (Long et al. 2022) and (Lin et al. 2021) to randomly select as few as 3 views for each object scan...' and 'Please refer to our supplementary materials for more details on the network training and coarse-to-fine learning strategy.' |
| Hardware Specification | No | No specific hardware details (e.g., GPU/CPU models, memory) used for running the experiments are provided. |
| Software Dependencies | No | No specific software dependencies with version numbers (e.g., libraries, frameworks) are mentioned. |
| Experiment Setup | No | No specific hyperparameters (e.g., learning rate, batch size, number of epochs, optimizer settings) or system-level training configurations are provided in the main text. Details are referred to supplementary materials. |