{"url":"/method/seam","slug":"seam","name":"SEAM","full_name":"Self-supervised Equivariant Attention Mechanism","full_name_withheld":false,"description_markdown":"**Self-supervised Equivariant Attention Mechanism**, or **SEAM**, is an attention mechanism for weakly supervised semantic segmentation. The SEAM applies consistency regularization on CAMs from various transformed images to provide self-supervision for network learning. To further improve the network prediction consistency, SEAM introduces the pixel correlation module (PCM), which captures context appearance information for each pixel and revises original CAMs by learned affinity attention maps. The SEAM is implemented by a [siamese network](https://paperswithcode.com/method/siamese-network) with equivariant cross regularization (ECR) loss, which regularizes the original CAMs and the revised CAMs on different branches.","description_state":"present","introduced_year":null,"introduced_by":{"title":"Self-supervised Equivariant Attention Mechanism for Weakly Supervised Semantic Segmentation","paper":"/paper/self-supervised-equivariant-attention","first_author":"Yude Wang","n_authors":5,"url_abs":null,"archive_paper_url":"https://paperswithcode.com/paper/self-supervised-equivariant-attention"},"source":{"url":"https://arxiv.org/abs/2004.04581v1","title":"Self-supervised Equivariant Attention Mechanism for Weakly Supervised Semantic Segmentation","url_on_a_paper_host":true},"code_snippet_url":null,"code_snippet_url_on_a_code_host":false,"categories":[{"area":"General","area_id":"general","collection":"Attention Mechanisms","url":"/methods/category/attention-mechanisms","pwc_aliases":["attention-mechanisms-1"]}],"n_papers_tagged":14,"archive_num_papers":14,"papers_newest_first":[{"paper":"/paper/self-destructive-language-model","title":"Self-Destructive Language Model","date":"2025-05-18","arxiv_id":"2505.12186","n_code_links":0,"syntology":{"ran":0,"of":6,"unverified":6,"pointer_only":0}},{"paper":null,"title":"SEAM: A Stochastic Benchmark for Multi-Document Tasks","date":"2024-06-23","arxiv_id":"2406.16086","n_code_links":0,"syntology":null},{"paper":"/paper/it-takes-two-on-the-seamlessness-between","title":"It Takes Two: On the Seamlessness between Reward and Policy Model in RLHF","date":"2024-06-12","arxiv_id":"2406.07971","n_code_links":1,"syntology":{"ran":12,"of":12,"unverified":0,"pointer_only":12}},{"paper":null,"title":"CVFC: Attention-Based Cross-View Feature Consistency for Weakly Supervised Semantic Segmentation of Pathology Images","date":"2023-08-21","arxiv_id":"2308.10449","n_code_links":0,"syntology":null},{"paper":null,"title":"Joint Microseismic Event Detection and Location with a Detection Transformer","date":"2023-07-16","arxiv_id":"2307.09207","n_code_links":0,"syntology":null},{"paper":"/paper/high-fidelity-pseudo-labels-for-boosting","title":"High-fidelity Pseudo-labels for Boosting Weakly-Supervised Segmentation","date":"2023-04-05","arxiv_id":"2304.02621","n_code_links":1,"syntology":null},{"paper":null,"title":"ContrasInver: Ultra-Sparse Label Semi-supervised Regression for Multi-dimensional Seismic Inversion","date":"2023-02-13","arxiv_id":"2302.06441","n_code_links":0,"syntology":null},{"paper":null,"title":"Selective Amnesia: On Efficient, High-Fidelity and Blind Suppression of Backdoor Effects in Trojaned Machine Learning Models","date":"2022-12-09","arxiv_id":"2212.04687","n_code_links":0,"syntology":null},{"paper":null,"title":"Out-of-Candidate Rectification for Weakly Supervised Semantic Segmentation","date":"2022-11-22","arxiv_id":"2211.12268","n_code_links":0,"syntology":null},{"paper":null,"title":"CONSS: Contrastive Learning Approach for Semi-Supervised Seismic Facies Classification","date":"2022-10-10","arxiv_id":"2210.04776","n_code_links":0,"syntology":null},{"paper":"/paper/yolo-facev2-a-scale-and-occlusion-aware-face","title":"YOLO-FaceV2: A Scale and Occlusion Aware Face Detector","date":"2022-08-03","arxiv_id":"2208.02019","n_code_links":1,"syntology":null},{"paper":"/paper/semi-supervised-impedance-inversion-by","title":"Semi-supervised Impedance Inversion by Bayesian Neural Network Based on 2-d CNN Pre-training","date":"2021-11-20","arxiv_id":"2111.10596","n_code_links":1,"syntology":null},{"paper":"/paper/movingfashion-a-benchmark-for-the-video-to","title":"MovingFashion: a Benchmark for the Video-to-Shop Challenge","date":"2021-10-06","arxiv_id":"2110.02627","n_code_links":1,"syntology":null},{"paper":"/paper/self-supervised-equivariant-attention","title":"Self-supervised Equivariant Attention Mechanism for Weakly Supervised Semantic Segmentation","date":"2020-04-09","arxiv_id":"2004.04581","n_code_links":2,"syntology":{"ran":3,"of":3,"unverified":0,"pointer_only":2}}],"papers_shown":14,"tasks":[{"task":"/task/semantic-segmentation","name":"Semantic Segmentation","papers":4},{"task":"/task/weakly-supervised-semantic-segmentation-1","name":"Weakly supervised Semantic Segmentation","papers":4},{"task":"/task/weakly-supervised-semantic-segmentation","name":"Weakly-Supervised Semantic Segmentation","papers":4},{"task":"/task/contrastive-learning","name":"Contrastive Learning","papers":2},{"task":"/task/segmentation","name":"Segmentation","papers":2},{"task":"/task/bayesian-inference","name":"Bayesian Inference","papers":1},{"task":"/task/continual-learning","name":"Continual Learning","papers":1},{"task":"/task/coreference-resolution","name":"Coreference Resolution","papers":1},{"task":"/task/data-augmentation","name":"Data Augmentation","papers":1},{"task":"/task/decoder","name":"Decoder","papers":1},{"task":"/task/event-detection","name":"Event Detection","papers":1},{"task":"/task/face-detection","name":"Face Detection","papers":1},{"task":"/task/facies-classification","name":"Facies Classification","papers":1},{"task":"/task/image-classification","name":"Image Classification","papers":1},{"task":"/task/image-segmentation","name":"Image Segmentation","papers":1},{"task":"/task/language-modeling","name":"Language Modeling","papers":1},{"task":"/task/language-modelling","name":"Language Modelling","papers":1},{"task":"/task/multi-hop-question-answering","name":"Multi-hop Question Answering","papers":1},{"task":"/task/optical-character-recognition","name":"Optical Character Recognition (OCR)","papers":1},{"task":"/task/prognosis","name":"Prognosis","papers":1}],"tasks_shown":20,"n_tasks":31,"usage_by_year":[{"year":"2020","papers":1},{"year":"2021","papers":2},{"year":"2022","papers":4},{"year":"2023","papers":4},{"year":"2024","papers":2},{"year":"2025","papers":1}],"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/seam"},"syntology_read_at":"2026-09-24T18:15:14+00:00"}