Fully Attentional Network for Semantic Segmentation
Authors: Qi Song, Jie Li, Chenghong Li, Hao Guo, Rui Huang2280-2288
AAAI 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We have carried out extensive experiments on three challenging semantic segmentation datasets and our approach achieves state-of-the-art performance on these experiments. |
| Researcher Affiliation | Academia | 1The Chinese University of Hong Kong, Shenzhen 2Shenzhen Institute of Artificial Intelligence and Robotics for Society |
| Pseudocode | No | The paper describes the method and shows a diagram (Fig.3) but does not include a formal pseudocode or algorithm block. |
| Open Source Code | No | The paper does not explicitly state that the source code for the described methodology is available, nor does it provide any links to a code repository. |
| Open Datasets | Yes | To evaluate the proposed FLANet, we conduct extensive experiments on the Cityscapes (Cordts et al. 2016), the ADE20K (Zhou et al. 2017), and the PASCAL VOC (Everingham et al. 2009). |
| Dataset Splits | Yes | The 5K images with fine annotations are further divided into 2975, 500, and 1525 images for training, validation, and testing, respectively. (Cityscapes) The dataset contains 20K/2K images for training and validation which is densely labeled as 150 stuff/object categories. (ADE20K) The dataset contains 10582, 1449, 1456 images for training, validation, and testing. (PASCAL VOC) |
| Hardware Specification | No | The paper states 'the synchronized batch normalization is used to synchronize the mean and standard deviation of batch normalization across multiple GPUs' but does not specify any particular hardware model (e.g., GPU, CPU) or detailed computer specifications used for running the experiments. |
| Software Dependencies | No | The paper states 'Our implementation is based on Py Torch' but does not provide specific version numbers for PyTorch or any other software dependencies. |
| Experiment Setup | Yes | Our implementation is based on Py Torch, and uses Res Net101 and HRNet W48 pre-trained from Image Net (Russakovsky et al. 2015) as the backbone network. Following prior works (Yu et al. 2018), we apply the poly learning rate policy where the initial learning rate is multiplied by (1 iter max iter)0.9 after each iteration. Momentum and weight decay coefficients are set to 0.9 and 5e-4, respectively. All models are trained for 240 epochs with an initial learning rate of 1e-2 and batch size of 8. We set the crop size as 768 768 and 520 520 for Cityscapes and other datasets, respectively. For data augmentation, we apply the common color jittering, scaling (0.5 to 2.0), cropping, and flipping to augment the training data. Besides, the synchronized batch normalization is used to synchronize the mean and standard deviation of batch normalization across multiple GPUs. For evaluation, the commonly used Mean Io U metric is adopted. |