Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in [1].
Partial Wasserstein Adversarial Network for Non-rigid Point Set Registration
Authors: Ziming Wang, Nan Xue, Ling Lei, Gui-Song Xia
ICLR 2022 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We evaluate PWAN on practical point set registration tasks, and show that the proposed PWAN is robust, scalable and performs more favorably than the state-of-the-art methods. |
| Researcher Affiliation | Academia | Zi-Ming Wang, Nan Xue, Ling Lei, Gui-Song Xia CAPTAIN Wuhan University |
| Pseudocode | Yes | The detailed algorithm is presented in Alg. 1. |
| Open Source Code | No | The paper does not contain any statement explicitly releasing its source code or provide a link to a code repository. |
| Open Datasets | Yes | The bunny and armadillo datasets are from the Stanford Repository (Standford), and the monkey dataset is from the internet. [...] human face dataset (Zhang et al., 2008) and a 3D human dataset (Data Set). |
| Dataset Splits | No | The paper describes the generation of artificial datasets and evaluation procedures, but it does not specify explicit training, validation, and test splits (e.g., percentages or sample counts) for model training or hyperparameter tuning. |
| Hardware Specification | Yes | We benchmark the computation time of different methods on a computer with two Nvidia GTX TITAN GPUs and an Intel i7 CPU. |
| Software Dependencies | No | The paper mentions using 'Adam optimizer (Kingma & Ba, 2014)' and 'RMSprop optimizer (Tieleman & Hinton, 2012)', but it does not specify version numbers for these optimizers or for any other software libraries or programming languages used (e.g., Python, PyTorch). |
| Experiment Setup | Yes | We use a 5-layer point-wise multi-layer perception as our network (Fig. 10 in the appendix), and the parameters used in the experiments are given in Appx. F.5. [...] The learning rates of both optimizers are set to 10 4. For experiments in Sec. 5.3, the parameters as set as follows: For PWAN, we set (ρ, λ, σ, T) = (2, 0.01, 0.1, 2000). |