OctField: Hierarchical Implicit Functions for 3D Modeling

Authors: Jia-Heng Tang, Weikai Chen, jie Yang, Bo Wang, Songrun Liu, Bo Yang, Lin Gao

NeurIPS 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental In the section, we will first introduce our data preparation process and then evaluate our approach in a variety of applications, including shape reconstruction, shape generation and interpolation, scene reconstruction and shape completion. We also provide ablation study, more comparisons and implementation details in the supplemental materials.
Researcher Affiliation Collaboration Jia-Heng Tang 1,2, Weikai Chen 3, Jie Yang1,2, Bo Wang3, Songrun Liu3, Bo Yang3, and Lin Gao (B) 1,2 1Beijing Key Laboratory of Mobile Computing and Pervasive Device, Institute of Computing Technology, Chinese Academy of Sciences 2University of Chinese Academy of Sciences 3Tencent Games Digital Content Technology Center
Pseudocode No The paper describes the proposed method and network architecture in detail but does not include any explicitly labeled pseudocode or algorithm blocks.
Open Source Code No The paper does not provide an explicit statement about releasing its source code or a link to a code repository for the methodology described.
Open Datasets Yes Our network is trained and evaluated on the five biggest and commonly used object categories in the Shape Net dataset [5]: chair, table, airplane, car, and sofa.
Dataset Splits No For fair comparison, we use the officially released training and testing data splits. This statement mentions training and testing but does not provide specific details (percentages, counts) for a validation split or how the data is partitioned overall for reproducibility.
Hardware Specification No The paper does not provide specific hardware details such as GPU or CPU models, memory, or cloud instance types used for running its experiments.
Software Dependencies No The paper mentions using 'the voxelization code provided by [25]' but does not specify any other software dependencies with version numbers (e.g., programming languages, libraries, frameworks).
Experiment Setup Yes We set λ = 10.0, β = 0.01 throughout our experiments.