Reduced Representation of Deformation Fields for Effective Non-rigid Shape Matching

Authors: Ramana Subramanyam Sundararaman, Riccardo Marin, Emanuele Rodolà, Maks Ovsjanikov

NeurIPS 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We illustrate its effectiveness through stateof-the-art results across multiple deformable shape matching benchmarks. Our code and data are publicly available at: https://github.com/Sentient07/ Deformation Basis.
Researcher Affiliation Academia Ramana Sundararaman1, Riccardo Marin2,3, Emanuele Rodolà3, and Maks Ovsjanikov1 1LIX, Ecole Polytechnique, IP Paris 2University of Tübingen 3Sapienza University of Rome
Pseudocode No No structured pseudocode or algorithm blocks are provided. The method is described using text and mathematical equations.
Open Source Code Yes Our code and data are publicly available at: https://github.com/Sentient07/ Deformation Basis.
Open Datasets Yes Our code and data are publicly available at: https://github.com/Sentient07/ Deformation Basis.
Dataset Splits Yes We train our method on a subset of 1000 SURREAL shapes [80] for 1000 epochs with data-augmentation along Y-axis.
Hardware Specification No The main paper does not provide specific hardware details (GPU/CPU models, memory amounts, or detailed computer specifications) used for running its experiments. The ethics checklist mentions details are in the supplementary, but the prompt asks for main paper body.
Software Dependencies No The paper does not provide specific ancillary software details (e.g., library or solver names with version numbers).
Experiment Setup Yes We train our method on a subset of 1000 SURREAL shapes [80] for 1000 epochs with data-augmentation along Y-axis.