{"url":"/method/spatial-attention-guided-mask","slug":"spatial-attention-guided-mask","name":"Spatial Attention-Guided Mask","full_name":"Spatial Attention-Guided Mask","full_name_withheld":false,"description_markdown":"**A Spatial Attention-Guided Mask** is a module for [instance segmentation](https://paperswithcode.com/task/instance-segmentation) that predicts a segmentation mask on each detected box with a spatial attention map that helps to focus on informative pixels and suppress noise. The goal is to guide the mask head for spotlighting meaningful pixels and repressing uninformative ones. \r\n\r\nOnce features inside the predicted RoIs are extracted by [RoIAlign](https://paperswithcode.com/method/roi-align) with 14×14 resolution, those features are fed into four conv layers and the [spatial attention module](https://paperswithcode.com/method/spatial-attention-module) (SAM) sequentially. To exploit the spatial attention map $A\\_{sag}\\left(X\\_{i}\\right) \\in \\mathcal{R}^{1\\times{W}\\times{H}}$ as a feature descriptor given input feature map $X\\_{i} \\in \\mathcal{R}^{C×W×H}$, the SAM first generates pooled features $P\\_{avg}, P\\_{max} \\in \\mathcal{R}^{1\\times{W}\\times{H}}$ by both average and [max pooling](https://paperswithcode.com/method/max-pooling) operations respectively along the channel axis and aggregates them via concatenation. Then it is followed by a 3 × 3 conv layer and normalized by the sigmoid function. The computation process\r\nis summarized as follow:\r\n\r\n$$\r\nA\\_{sag}\\left(X\\_{i}\\right) = \\sigma\\left(F\\_{3\\times{3}}(P\\_{max} \\cdot P\\_{avg})\\right)\r\n$$\r\n\r\nwhere $\\sigma$ denotes the sigmoid function, $F\\_{3\\times{3}}$ is 3 × 3 conv layer and $\\cdot$ represents the concatenate operation. Finally, the attention guided feature map $X\\_{sag} ∈ \\mathcal{R}^{C\\times{W}\\times{H}}$ is computed as:\r\n\r\n$$\r\nX\\_{sag} = A\\_{sag}\\left(X\\_{i}\\right) \\otimes X\\_{i}\r\n$$\r\n\r\nwhere ⊗ denotes element-wise multiplication. After then, a 2 × 2 deconv upsamples the spatially attended feature map to 28 × 28 resolution. Lastly, a 1 × 1 conv is applied for predicting class-specific masks.","description_state":"present","introduced_year":null,"introduced_by":{"title":"CenterMask : Real-Time Anchor-Free Instance Segmentation","paper":"/paper/centermask-real-time-anchor-free-instance-1","first_author":"Youngwan Lee","n_authors":2,"url_abs":null,"archive_paper_url":"https://paperswithcode.com/paper/centermask-real-time-anchor-free-instance-1"},"source":{"url":"https://arxiv.org/abs/1911.06667v6","title":"CenterMask : Real-Time Anchor-Free Instance Segmentation","url_on_a_paper_host":true},"code_snippet_url":"https://github.com/youngwanLEE/CenterMask/blob/4a7a421913ecd57ddba23b27410d0a2f14565f33/maskrcnn_benchmark/modeling/make_layers.py#L199","code_snippet_url_on_a_code_host":true,"categories":[{"area":"General","area_id":"general","collection":"Attention Modules","url":"/methods/category/attention-modules","pwc_aliases":[]},{"area":"Computer Vision","area_id":"computer-vision","collection":"Mask Branches","url":"/methods/category/mask-branches","pwc_aliases":[]}],"n_papers_tagged":7,"archive_num_papers":7,"papers_newest_first":[{"paper":"/paper/polite-teacher-semi-supervised-instance","title":"Polite Teacher: Semi-Supervised Instance Segmentation with Mutual Learning and Pseudo-Label Thresholding","date":"2022-11-07","arxiv_id":"2211.03850","n_code_links":0,"syntology":null},{"paper":null,"title":"Intelligent detect for substation insulator defects based on CenterMask","date":"2022-08-31","arxiv_id":"2208.14598","n_code_links":0,"syntology":null},{"paper":null,"title":"CenterMask: single shot instance segmentation with point representation","date":"2020-04-09","arxiv_id":"2004.04446","n_code_links":0,"syntology":null},{"paper":"/paper/sa-unet-spatial-attention-u-net-for-retinal","title":"SA-UNet: Spatial Attention U-Net for Retinal Vessel Segmentation","date":"2020-04-07","arxiv_id":"2004.03696","n_code_links":4,"syntology":null},{"paper":null,"title":"Learning Oracle Attention for High-fidelity Face Completion","date":"2020-03-31","arxiv_id":"2003.13903","n_code_links":0,"syntology":null},{"paper":"/paper/context-aware-domain-adaptation-in-semantic","title":"Context-Aware Domain Adaptation in Semantic Segmentation","date":"2020-03-09","arxiv_id":"2003.04010","n_code_links":0,"syntology":null},{"paper":"/paper/centermask-real-time-anchor-free-instance-1","title":"CenterMask : Real-Time Anchor-Free Instance Segmentation","date":"2019-11-15","arxiv_id":"1911.06667","n_code_links":8,"syntology":{"ran":1,"of":2,"unverified":1,"pointer_only":2}}],"papers_shown":7,"tasks":[{"task":"/task/semantic-segmentation","name":"Semantic Segmentation","papers":4},{"task":"/task/instance-segmentation","name":"Instance Segmentation","papers":3},{"task":"/task/segmentation","name":"Segmentation","papers":3},{"task":"/task/semi-supervised-instance-segmentation","name":"Semi-Supervised Instance Segmentation","papers":2},{"task":"/task/data-augmentation","name":"Data Augmentation","papers":1},{"task":"/task/defect-detection","name":"Defect Detection","papers":1},{"task":"/task/domain-adaptation","name":"Domain Adaptation","papers":1},{"task":"/task/facial-inpainting","name":"Facial Inpainting","papers":1},{"task":"/task/object","name":"Object","papers":1},{"task":"/task/object-detection","name":"Object Detection","papers":1},{"task":"/task/panoptic-segmentation","name":"Panoptic Segmentation","papers":1},{"task":"/task/pseudo-label","name":"Pseudo Label","papers":1},{"task":"/task/real-time-object-detection","name":"Real-Time Object Detection","papers":1},{"task":"/task/real-time-instance-segmentation","name":"Real-time Instance Segmentation","papers":1},{"task":"/task/retinal-vessel-segmentation","name":"Retinal Vessel Segmentation","papers":1},{"task":"/task/unsupervised-domain-adaptation","name":"Unsupervised Domain Adaptation","papers":1},{"task":"/task/high","name":"Vocal Bursts Intensity Prediction","papers":1}],"tasks_shown":17,"n_tasks":17,"usage_by_year":[{"year":"2019","papers":1},{"year":"2020","papers":4},{"year":"2022","papers":2}],"row_source":"methods_table","archive":{"source":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","archive_url":"https://paperswithcode.com/method/spatial-attention-guided-mask"},"syntology_read_at":"2026-09-24T18:15:14+00:00"}