{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/sum-saliency-unification-through-mamba-for","title":"SUM: Saliency Unification through Mamba for Visual Attention Modeling","arxiv_id":"2406.17815","date":"2024-06-25","proceeding":null,"authors":["Alireza Hosseini","Amirhossein Kazerouni","Saeed Akhavan","Michael Brudno","Babak Taati"],"abstract":"Visual attention modeling, important for interpreting and prioritizing visual stimuli, plays a significant role in applications such as marketing, multimedia, and robotics. Traditional saliency prediction models, especially those based on Convolutional Neural Networks (CNNs) or Transformers, achieve notable success by leveraging large-scale annotated datasets. However, the current state-of-the-art (SOTA) models that use Transformers are computationally expensive. Additionally, separate models are often required for each image type, lacking a unified approach. In this paper, we propose Saliency Unification through Mamba (SUM), a novel approach that integrates the efficient long-range dependency modeling of Mamba with U-Net to provide a unified model for diverse image types. Using a novel Conditional Visual State Space (C-VSS) block, SUM dynamically adapts to various image types, including natural scenes, web pages, and commercial imagery, ensuring universal applicability across different data types. Our comprehensive evaluations across five benchmarks demonstrate that SUM seamlessly adapts to different visual characteristics and consistently outperforms existing models. These results position SUM as a versatile and powerful tool for advancing visual attention modeling, offering a robust solution universally applicable across different types of visual content.","url_abs":"https://arxiv.org/abs/2406.17815v2","url_pdf":"https://arxiv.org/pdf/2406.17815v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"sum-saliency-unification-through-mamba-for","repo_url":"https://github.com/Arhosseini77/SUM","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"mamba","task_name":"Mamba"},{"task_slug":"marketing","task_name":"Marketing"},{"task_slug":"saliency-detection","task_name":"Saliency Detection"},{"task_slug":"saliency-prediction","task_name":"Saliency Prediction"}],"methods":[{"method_slug":"attention","method_name":"Attention"},{"method_slug":"concatenated-skip-connection","method_name":"Concatenated Skip Connection"},{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"mamba","method_name":"Mamba"},{"method_slug":"max-pooling","method_name":"Max Pooling"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"relu","method_name":"ReLU"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"stochastic-depth","method_name":"Stochastic Depth"},{"method_slug":"swin-transformer","method_name":"Swin Transformer"},{"method_slug":"u-net","method_name":"U-Net"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/saliency-detection-on-cat2000","task":"Saliency Detection","dataset":"CAT2000","model":"SUM","rank_in_archive_order":1,"of":2,"metrics":{"AUC":"0.888","NSS":"2.423"},"uses_additional_data":false},{"leaderboard":"/sota/saliency-prediction-on-cat2000","task":"Saliency Prediction","dataset":"CAT2000","model":"SUM","rank_in_archive_order":1,"of":1,"metrics":{"KL":"0.27"},"uses_additional_data":false},{"leaderboard":"/sota/saliency-prediction-on-mit300","task":"Saliency Prediction","dataset":"MIT300","model":"SUM","rank_in_archive_order":1,"of":2,"metrics":{"AUC-Judd":"0.913","CC":"0.768","KLD":"0.563","NSS":"2.839","SIM":"0.63"},"uses_additional_data":false},{"leaderboard":"/sota/saliency-prediction-on-saleci","task":"Saliency Prediction","dataset":"SALECI","model":"SUM","rank_in_archive_order":1,"of":5,"metrics":{"KL":"0.473"},"uses_additional_data":false},{"leaderboard":"/sota/saliency-prediction-on-salicon","task":"Saliency Prediction","dataset":"SALICON","model":"SUM","rank_in_archive_order":1,"of":5,"metrics":{"AUC":"0.876","CC":"0.909","KLD":"0.192","NSS":"1.981","SIM":"0.804"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/2406.17815","atlas_url":"https://app.syntology.ai/?focus=2406.17815","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2406.17815"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/Arhosseini77/SUM","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran":14},"by_repo_kind":{"official":{"samples":14,"ran":14,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"c00dcf8428ee981b","entry":"flops_selective_scan_ref","repo":"Arhosseini77/SUM","repo_kind":"official","path":"net/models/vmamba.py","file_url":"https://github.com/Arhosseini77/SUM/blob/HEAD/net/models/vmamba.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"c00dcf8428ee981b"}},{"code_sha256_prefix":"23a1f34627822bb3","entry":"get_logger","repo":"Arhosseini77/SUM","repo_kind":"official","path":"net/utils.py","file_url":"https://github.com/Arhosseini77/SUM/blob/HEAD/net/utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"23a1f34627822bb3"}},{"code_sha256_prefix":"ae1a2dd9d62e7fad","entry":"get_optimizer","repo":"Arhosseini77/SUM","repo_kind":"official","path":"net/utils.py","file_url":"https://github.com/Arhosseini77/SUM/blob/HEAD/net/utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"ae1a2dd9d62e7fad"}},{"code_sha256_prefix":"dea42821258184c2","entry":"get_scheduler","repo":"Arhosseini77/SUM","repo_kind":"official","path":"net/utils.py","file_url":"https://github.com/Arhosseini77/SUM/blob/HEAD/net/utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"dea42821258184c2"}},{"code_sha256_prefix":"a214ec961739db31","entry":"load_and_preprocess_image","repo":"Arhosseini77/SUM","repo_kind":"official","path":"net/inference_utils.py","file_url":"https://github.com/Arhosseini77/SUM/blob/HEAD/net/inference_utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"a214ec961739db31"}},{"code_sha256_prefix":"774d792a6525842a","entry":"loss_CC","repo":"Arhosseini77/SUM","repo_kind":"official","path":"utils/loss_function.py","file_url":"https://github.com/Arhosseini77/SUM/blob/HEAD/utils/loss_function.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"774d792a6525842a"}},{"code_sha256_prefix":"e9183c75893d49f0","entry":"loss_KLdiv","repo":"Arhosseini77/SUM","repo_kind":"official","path":"utils/loss_function.py","file_url":"https://github.com/Arhosseini77/SUM/blob/HEAD/utils/loss_function.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"e9183c75893d49f0"}},{"code_sha256_prefix":"9596d79dd2070fff","entry":"loss_similarity","repo":"Arhosseini77/SUM","repo_kind":"official","path":"utils/loss_function.py","file_url":"https://github.com/Arhosseini77/SUM/blob/HEAD/utils/loss_function.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"9596d79dd2070fff"}},{"code_sha256_prefix":"f0c787b1471d4421","entry":"mean_std","repo":"Arhosseini77/SUM","repo_kind":"official","path":"optuna_network.py","file_url":"https://github.com/Arhosseini77/SUM/blob/HEAD/optuna_network.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"f0c787b1471d4421"}},{"code_sha256_prefix":"95949c1af8f08e9d","entry":"modulate","repo":"Arhosseini77/SUM","repo_kind":"official","path":"net/models/vmamba.py","file_url":"https://github.com/Arhosseini77/SUM/blob/HEAD/net/models/vmamba.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"95949c1af8f08e9d"}},{"code_sha256_prefix":"e26191359b0c82c3","entry":"modulate_scale","repo":"Arhosseini77/SUM","repo_kind":"official","path":"net/models/vmamba.py","file_url":"https://github.com/Arhosseini77/SUM/blob/HEAD/net/models/vmamba.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"e26191359b0c82c3"}},{"code_sha256_prefix":"ebfa76e7f72447ce","entry":"postprocess_img","repo":"Arhosseini77/SUM","repo_kind":"official","path":"utils/data_process_uni.py","file_url":"https://github.com/Arhosseini77/SUM/blob/HEAD/utils/data_process_uni.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"ebfa76e7f72447ce"}},{"code_sha256_prefix":"40e257a513360d15","entry":"predict_saliency_map","repo":"Arhosseini77/SUM","repo_kind":"official","path":"net/inference_utils.py","file_url":"https://github.com/Arhosseini77/SUM/blob/HEAD/net/inference_utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"40e257a513360d15"}},{"code_sha256_prefix":"f3af41b9e983a72c","entry":"preprocess_img","repo":"Arhosseini77/SUM","repo_kind":"official","path":"utils/data_process_uni.py","file_url":"https://github.com/Arhosseini77/SUM/blob/HEAD/utils/data_process_uni.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"f3af41b9e983a72c"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}