{"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/joint-maximum-purity-forest-with-application","title":"Joint Maximum Purity Forest with Application to Image Super-Resolution","arxiv_id":"1708.09200","date":"2017-08-30","proceeding":null,"authors":["Hailiang Li","Kin-Man Lam","Dong Li"],"abstract":"In this paper, we propose a novel random-forest scheme, namely Joint Maximum\nPurity Forest (JMPF), for classification, clustering, and regression tasks. In\nthe JMPF scheme, the original feature space is transformed into a compactly\npre-clustered feature space, via a trained rotation matrix. The rotation matrix\nis obtained through an iterative quantization process, where the input data\nbelonging to different classes are clustered to the respective vertices of the\nnew feature space with maximum purity. In the new feature space, orthogonal\nhyperplanes, which are employed at the split-nodes of decision trees in random\nforests, can tackle the clustering problems effectively. We evaluated our\nproposed method on public benchmark datasets for regression and classification\ntasks, and experiments showed that JMPF remarkably outperforms other\nstate-of-the-art random-forest-based approaches. Furthermore, we applied JMPF\nto image super-resolution, because the transformed, compact features are more\ndiscriminative to the clustering-regression scheme. Experiment results on\nseveral public benchmark datasets also showed that the JMPF-based image\nsuper-resolution scheme is consistently superior to recent state-of-the-art\nimage super-resolution algorithms.","url_abs":"http://arxiv.org/abs/1708.09200v1","url_pdf":"http://arxiv.org/pdf/1708.09200v1.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":"joint-maximum-purity-forest-with-application","repo_url":"https://github.com/HarleyHK/JMPF","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"clustering","task_name":"Clustering"},{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"image-super-resolution","task_name":"Image Super-Resolution"},{"task_slug":"quantization","task_name":"Quantization"},{"task_slug":"super-resolution","task_name":"Super-Resolution"},{"task_slug":"regression-1","task_name":"regression"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/image-super-resolution-on-bsd100-4x-upscaling","task":"Image Super-Resolution","dataset":"BSD100 - 4x upscaling","model":"JMPF+","rank_in_archive_order":56,"of":71,"metrics":{"PSNR":"26.87"},"uses_additional_data":false},{"leaderboard":"/sota/image-super-resolution-on-set14-4x-upscaling","task":"Image Super-Resolution","dataset":"Set14 - 4x upscaling","model":"JMPF+","rank_in_archive_order":96,"of":104,"metrics":{"PSNR":"27.37"},"uses_additional_data":false}],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}