{"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/hi-fi-hierarchical-feature-integration-for","title":"Hi-Fi: Hierarchical Feature Integration for Skeleton Detection","arxiv_id":"1801.01849","date":"2018-01-05","proceeding":null,"authors":["Kai Zhao","Wei Shen","Shang-Hua Gao","Dandan Li","Ming-Ming Cheng"],"abstract":"In natural images, the scales (thickness) of object skeletons may\ndramatically vary among objects and object parts, making object skeleton\ndetection a challenging problem. We present a new convolutional neural network\n(CNN) architecture by introducing a novel hierarchical feature integration\nmechanism, named Hi-Fi, to address the skeleton detection problem. The proposed\nCNN-based approach has a powerful multi-scale feature integration ability that\nintrinsically captures high-level semantics from deeper layers as well as\nlow-level details from shallower layers. % By hierarchically integrating\ndifferent CNN feature levels with bidirectional guidance, our approach (1)\nenables mutual refinement across features of different levels, and (2)\npossesses the strong ability to capture both rich object context and\nhigh-resolution details. Experimental results show that our method\nsignificantly outperforms the state-of-the-art methods in terms of effectively\nfusing features from very different scales, as evidenced by a considerable\nperformance improvement on several benchmarks.","url_abs":"http://arxiv.org/abs/1801.01849v4","url_pdf":"http://arxiv.org/pdf/1801.01849v4.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":[],"tasks":[{"task_slug":"object","task_name":"Object"},{"task_slug":"object-skeleton-detection","task_name":"Object Skeleton Detection"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/object-skeleton-detection-on-sk-large","task":"Object Skeleton Detection","dataset":"SK-LARGE","model":"Hi-Fi","rank_in_archive_order":2,"of":2,"metrics":{"F-Measure":"0.724"},"uses_additional_data":false}],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}