Aligning Gradient and Hessian for Neural Signed Distance Function

Authors: Ruian Wang, Zixiong Wang, Yunxiao Zhang, Shuangmin Chen, Shiqing Xin, Changhe Tu, Wenping Wang

NeurIPS 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experimental results demonstrate its ability to accurately recover the underlying shape while effectively suppressing the presence of ghost geometry.
Researcher Affiliation Academia Ruian Wang Shandong University wra.time@gmail.com; Zixiong Wang Nankai University zixiong_wang@outlook.com; Yunxiao Zhang Shandong University zhangyunxiaox@gmail.com; Shuangmin Chen Qingdao University of Science and Technology csmqq@163.com; Shiqing Xin Shan Dong University xinshiqing@sdu.edu.cn; Changhe Tu Shandong University chtu@sdu.edu.cn; Wenping Wang TEXAS A&M UNIVERSITY wenping@cs.hku.hk
Pseudocode No The paper does not contain structured pseudocode or algorithm blocks, nor does it include a figure or section explicitly labeled 'Pseudocode' or 'Algorithm'.
Open Source Code No The paper does not include an unambiguous statement about releasing code for the work described, nor does it provide a direct link to a source-code repository.
Open Datasets Yes The Surface Reconstruction Benchmark (SRB) [49]... The Shape Net dataset [13]... The ABC dataset [24]... Thingi10K [57]... 3D Scene dataset [56]... The D-Faust [7] dataset...
Dataset Splits No The paper mentions splitting for training and testing on datasets like ShapeNet and D-Faust (e.g., '6K scans are used for training and 2K scans for testing' for D-Faust), but it does not explicitly provide details about a separate validation set split or its size/methodology.
Hardware Specification No The paper does not provide specific hardware details such as exact GPU/CPU models, processor types, or memory amounts used for running its experiments. It only mentions general experimental setups without hardware specifications.
Software Dependencies No The paper mentions leveraging 'SIREN [40]' and 'Neural-Pull [5]' but does not provide specific version numbers for these or any other software dependencies, such as programming languages or libraries.
Experiment Setup Yes we set δ = 10 by default. At the same time, we set the default value of α to 6. More details can be checked in our supplementary material.