Structure-Aware Image Segmentation with Homotopy Warping
Authors: Xiaoling Hu
NeurIPS 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We conduct extensive experiments to demonstrate the effectiveness of the proposed method. |
| Researcher Affiliation | Academia | Xiaoling Hu Department of Computer Science Stony Brook University xiaolhu@cs.stonybrook.edu |
| Pseudocode | No | The paper describes the proposed algorithm in prose (Section 3.4) but does not provide a formal pseudocode block or algorithm figure. |
| Open Source Code | Yes | 3. If you ran experiments... (a) Did you include the code, data, and instructions needed to reproduce the main experimental results (either in the supplemental material or as a URL)? [Yes] See the Supplemental Material. |
| Open Datasets | Yes | Specifically, we use four natural and biomedical 2D datasets (Road Tracer [4], Deep Globe [12], Mass. [34], DRIVE [44]) and one more 3D biomedical dataset (CREMI 2) to validate the efficacy of the propose method. More details about the datasets and the split of the training and validation subsets are included in Sec. A.6. 2https://cremi.org/ |
| Dataset Splits | Yes | More details about the datasets and the split of the training and validation subsets are included in Sec. A.6. |
| Hardware Specification | Yes | All the experiments are performed on a Tesla V100-SXM2 GPU (32G Memory), and an Intel(R) Xeon(R) Gold 6140 CPU@2.30 GHz. |
| Software Dependencies | Yes | We use Py Torch framework (Version: 1.7.1) to implement the proposed method. |
| Experiment Setup | Yes | For 2D datasets, the batch size is set as 16, and the initial learning rate is 0.01. We randomly crop patches with the size of 512 512 and then feed them into the 2D UNet. For 3D case, the batch size is also 16, while the input size is 128 128 16. |