{"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/deep-reasoning-with-knowledge-graph-for","title":"Deep Reasoning with Knowledge Graph for Social Relationship Understanding","arxiv_id":"1807.00504","date":"2018-07-02","proceeding":null,"authors":["Zhouxia Wang","Tianshui Chen","Jimmy Ren","Weihao Yu","Hui Cheng","Liang Lin"],"abstract":"Social relationships (e.g., friends, couple etc.) form the basis of the\nsocial network in our daily life. Automatically interpreting such relationships\nbears a great potential for the intelligent systems to understand human\nbehavior in depth and to better interact with people at a social level. Human\nbeings interpret the social relationships within a group not only based on the\npeople alone, and the interplay between such social relationships and the\ncontextual information around the people also plays a significant role.\nHowever, these additional cues are largely overlooked by the previous studies.\nWe found that the interplay between these two factors can be effectively\nmodeled by a novel structured knowledge graph with proper message propagation\nand attention. And this structured knowledge can be efficiently integrated into\nthe deep neural network architecture to promote social relationship\nunderstanding by an end-to-end trainable Graph Reasoning Model (GRM), in which\na propagation mechanism is learned to propagate node message through the graph\nto explore the interaction between persons of interest and the contextual\nobjects. Meanwhile, a graph attentional mechanism is introduced to explicitly\nreason about the discriminative objects to promote recognition. Extensive\nexperiments on the public benchmarks demonstrate the superiority of our method\nover the existing leading competitors.","url_abs":"http://arxiv.org/abs/1807.00504v1","url_pdf":"http://arxiv.org/pdf/1807.00504v1.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":"deep-reasoning-with-knowledge-graph-for","repo_url":"https://github.com/HCPLab-SYSU/SR","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"visual-social-relationship-recognition","task_name":"Visual Social Relationship Recognition"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/visual-social-relationship-recognition-on-1","task":"Visual Social Relationship Recognition","dataset":"PIPA","model":"GRM","rank_in_archive_order":3,"of":6,"metrics":{"Accuracy":"62.3"},"uses_additional_data":false},{"leaderboard":"/sota/visual-social-relationship-recognition-on","task":"Visual Social Relationship Recognition","dataset":"PISC","model":"GRM","rank_in_archive_order":4,"of":5,"metrics":{"mAP":"68.7","mAP (Coarse)":"82.8"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1807.00504","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}