Amodal Segmentation Based on Visible Region Segmentation and Shape Prior
Authors: Yuting Xiao, Yanyu Xu, Ziming Zhong, Weixin Luo, Jiawei Li, Shenghua Gao2995-3003
AAAI 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Our proposed model is evaluated on three datasets. Experiments show that our proposed model outperforms existing state-of-the-art methods. |
| Researcher Affiliation | Collaboration | 1 Shanghai Tech University 2 Shanghai Engineering Research Center of Intelligent Vision and Imaging 3 Alibaba Group 4 Institute of High Performance Computing, A*STAR |
| Pseudocode | No | The paper does not contain any structured pseudocode or algorithm blocks. |
| Open Source Code | Yes | The code is available at https://github.com/Yuting Xiao/Amodal-Segmentation Based-on-Visible-Region-Segmentation-and-Shape-Prior. |
| Open Datasets | Yes | We evaluate the model performance for amodal segmentation on three datasets: the D2SA (D2S amodal) (Follmann et al. 2019), the KINS dataset (Qi et al. 2019), the COCOA cls dataset (Zhu et al. 2017). |
| Dataset Splits | Yes | The D2SA dataset... It contains 2000 images in the training set and 3600 images in the validation set...The KINS dataset... It consists of 7474 images in the training set and 7517 images in the validation set. The COCOA cls dataset... It consists of 2476 images in the training set and 1223 images in the validation set. |
| Hardware Specification | No | The paper does not provide specific hardware details (e.g., GPU/CPU models, memory) used for running its experiments. |
| Software Dependencies | No | The paper mentions using 'Detectron2 (Wu et al. 2019) on the Py Torch framework' but does not specify version numbers for these software components. |
| Experiment Setup | Yes | The main parameter setting is: For the D2SA dataset, batch size(2), learning rate (0.005), and the number of iteration (70000). For the KINS dataset, batch size(1), learning rate (0.0025), and the number of iteration (48000). For the COCOA cls dataset, batch size (2), learning rate (0.0005), and the number of iteration (10000). |