{"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/m-pact-an-open-source-platform-for-repeatable","title":"M-PACT: An Open Source Platform for Repeatable Activity Classification Research","arxiv_id":"1804.05879","date":"2018-04-16","proceeding":null,"authors":["Eric Hofesmann","Madan Ravi Ganesh","Jason J. Corso"],"abstract":"There are many hurdles that prevent the replication of existing work which\nhinders the development of new activity classification models. These hurdles\ninclude switching between multiple deep learning libraries and the development\nof boilerplate experimental pipelines. We present M-PACT to overcome existing\nissues by removing the need to develop boilerplate code which allows users to\nquickly prototype action classification models while leveraging existing\nstate-of-the-art (SOTA) models available in the platform. M-PACT is the first\nto offer four SOTA activity classification models, I3D, C3D, ResNet50+LSTM, and\nTSN, under a single platform with reproducible competitive results. This\nplatform allows for the generation of models and results over activity\nrecognition datasets through the use of modular code, various preprocessing and\nneural network layers, and seamless data flow. In this paper, we present the\nsystem architecture, detail the functions of various modules, and describe the\nbasic tools to develop a new model in M-PACT.","url_abs":"http://arxiv.org/abs/1804.05879v3","url_pdf":"http://arxiv.org/pdf/1804.05879v3.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":"m-pact-an-open-source-platform-for-repeatable","repo_url":"https://github.com/MichiganCOG/M-PACT","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"action-classification","task_name":"Action Classification"},{"task_slug":"activity-recognition","task_name":"Activity Recognition"},{"task_slug":"classification-1","task_name":"Classification"},{"task_slug":"classification","task_name":"General Classification"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}