{"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/sshmt-semi-supervised-hierarchical-merge-tree","title":"SSHMT: Semi-supervised Hierarchical Merge Tree for Electron Microscopy Image Segmentation","arxiv_id":"1608.04051","date":"2016-08-14","proceeding":null,"authors":["Ting Liu","Miaomiao Zhang","Mehran Javanmardi","Nisha Ramesh","Tolga Tasdizen"],"abstract":"Region-based methods have proven necessary for improving segmentation\naccuracy of neuronal structures in electron microscopy (EM) images. Most\nregion-based segmentation methods use a scoring function to determine region\nmerging. Such functions are usually learned with supervised algorithms that\ndemand considerable ground truth data, which are costly to collect. We propose\na semi-supervised approach that reduces this demand. Based on a merge tree\nstructure, we develop a differentiable unsupervised loss term that enforces\nconsistent predictions from the learned function. We then propose a Bayesian\nmodel that combines the supervised and the unsupervised information for\nprobabilistic learning. The experimental results on three EM data sets\ndemonstrate that by using a subset of only 3% to 7% of the entire ground truth\ndata, our approach consistently performs close to the state-of-the-art\nsupervised method with the full labeled data set, and significantly outperforms\nthe supervised method with the same labeled subset.","url_abs":"http://arxiv.org/abs/1608.04051v1","url_pdf":"http://arxiv.org/pdf/1608.04051v1.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":"sshmt-semi-supervised-hierarchical-merge-tree","repo_url":"https://github.com/tingliu/glia","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"electron-microscopy-image-segmentation","task_name":"Electron Microscopy Image Segmentation"},{"task_slug":"image-segmentation","task_name":"Image Segmentation"},{"task_slug":"segmentation","task_name":"Segmentation"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}