{"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/cogtree-cognition-tree-loss-for-unbiased","title":"CogTree: Cognition Tree Loss for Unbiased Scene Graph Generation","arxiv_id":"2009.07526","date":"2020-09-16","proceeding":null,"authors":["Jing Yu","Yuan Chai","Yujing Wang","Yue Hu","Qi Wu"],"abstract":"Scene graphs are semantic abstraction of images that encourage visual understanding and reasoning. However, the performance of Scene Graph Generation (SGG) is unsatisfactory when faced with biased data in real-world scenarios. Conventional debiasing research mainly studies from the view of balancing data distribution or learning unbiased models and representations, ignoring the correlations among the biased classes. In this work, we analyze this problem from a novel cognition perspective: automatically building a hierarchical cognitive structure from the biased predictions and navigating that hierarchy to locate the relationships, making the tail relationships receive more attention in a coarse-to-fine mode. To this end, we propose a novel debiasing Cognition Tree (CogTree) loss for unbiased SGG. We first build a cognitive structure CogTree to organize the relationships based on the prediction of a biased SGG model. The CogTree distinguishes remarkably different relationships at first and then focuses on a small portion of easily confused ones. Then, we propose a debiasing loss specially for this cognitive structure, which supports coarse-to-fine distinction for the correct relationships. The loss is model-agnostic and consistently boosting the performance of several state-of-the-art models. The code is available at: https://github.com/CYVincent/Scene-Graph-Transformer-CogTree.","url_abs":"https://arxiv.org/abs/2009.07526v2","url_pdf":"https://arxiv.org/pdf/2009.07526v2.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":"cogtree-cognition-tree-loss-for-unbiased","repo_url":"https://github.com/CYVincent/Scene-Graph-Transformer-CogTree","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"graph-generation","task_name":"Graph Generation"},{"task_slug":"scene-graph-generation","task_name":"Scene Graph Generation"},{"task_slug":"unbiased-scene-graph-generation","task_name":"Unbiased Scene Graph Generation"}],"methods":[{"method_slug":"absolute-position-encodings","method_name":"Absolute Position Encodings"},{"method_slug":"adam","method_name":"Adam"},{"method_slug":"attention","method_name":"Attention"},{"method_slug":"bpe","method_name":"BPE"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"label-smoothing","method_name":"Label Smoothing"},{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"position-wise-feed-forward-layer","method_name":"Position-Wise Feed-Forward Layer"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"transformer","method_name":"Transformer"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/scene-graph-generation-on-visual-genome","task":"Scene Graph Generation","dataset":"Visual Genome","model":"CogTree","rank_in_archive_order":16,"of":19,"metrics":{"mean Recall @20":"7.9"},"uses_additional_data":false},{"leaderboard":"/sota/unbiased-scene-graph-generation-on-visual","task":"Unbiased Scene Graph Generation","dataset":"Visual Genome","model":"CogTree (VCTree-ResNeXt-101-FPN backbone; PredCls mode)","rank_in_archive_order":26,"of":31,"metrics":{"F@100":"35.9","mR@20":"22.0"},"uses_additional_data":false},{"leaderboard":"/sota/unbiased-scene-graph-generation-on-visual","task":"Unbiased Scene Graph Generation","dataset":"Visual Genome","model":"CogTree (MOTIFS-ResNeXt-101-FPN backbone; PredCls mode)","rank_in_archive_order":27,"of":31,"metrics":{"F@100":"32.4","mR@20":"20.9"},"uses_additional_data":false},{"leaderboard":"/sota/unbiased-scene-graph-generation-on-visual","task":"Unbiased Scene Graph Generation","dataset":"Visual Genome","model":"CogTree (VCTree-ResNeXt-101-FPN backbone; SGCls mode)","rank_in_archive_order":28,"of":31,"metrics":{"mR@20":"15.4"},"uses_additional_data":false},{"leaderboard":"/sota/unbiased-scene-graph-generation-on-visual","task":"Unbiased Scene Graph Generation","dataset":"Visual Genome","model":"CogTree (MOTIFS-ResNeXt-101-FPN backbone; SGCls mode)","rank_in_archive_order":29,"of":31,"metrics":{"mR@20":"12.1"},"uses_additional_data":false},{"leaderboard":"/sota/unbiased-scene-graph-generation-on-visual","task":"Unbiased Scene Graph Generation","dataset":"Visual Genome","model":"CogTree (MOTIFS-ResNeXt-101-FPN backbone; SGDet mode)","rank_in_archive_order":30,"of":31,"metrics":{"mR@20":"7.9"},"uses_additional_data":false},{"leaderboard":"/sota/unbiased-scene-graph-generation-on-visual","task":"Unbiased Scene Graph Generation","dataset":"Visual Genome","model":"CogTree (VCTree-ResNeXt-101-FPN backbone; SGDet mode)","rank_in_archive_order":31,"of":31,"metrics":{"mR@20":"7.8"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/2009.07526","atlas_url":"https://app.syntology.ai/?focus=2009.07526","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}