{"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/adaptive-feature-processing-for-robust-human","title":"Adaptive Feature Processing for Robust Human Activity Recognition on a Novel Multi-Modal Dataset","arxiv_id":"1901.02858","date":"2019-01-09","proceeding":null,"authors":["Mirco Moencks","Varuna De Silva","Jamie Roche","Ahmet Kondoz"],"abstract":"Human Activity Recognition (HAR) is a key building block of many emerging\napplications such as intelligent mobility, sports analytics, ambient-assisted\nliving and human-robot interaction. With robust HAR, systems will become more\nhuman-aware, leading towards much safer and empathetic autonomous systems.\nWhile human pose detection has made significant progress with the dawn of deep\nconvolutional neural networks (CNNs), the state-of-the-art research has almost\nexclusively focused on a single sensing modality, especially video. However, in\nsafety critical applications it is imperative to utilize multiple sensor\nmodalities for robust operation. To exploit the benefits of state-of-the-art\nmachine learning techniques for HAR, it is extremely important to have\nmultimodal datasets. In this paper, we present a novel, multi-modal sensor\ndataset that encompasses nine indoor activities, performed by 16 participants,\nand captured by four types of sensors that are commonly used in indoor\napplications and autonomous vehicles. This multimodal dataset is the first of\nits kind to be made openly available and can be exploited for many applications\nthat require HAR, including sports analytics, healthcare assistance and indoor\nintelligent mobility. We propose a novel data preprocessing algorithm to enable\nadaptive feature extraction from the dataset to be utilized by different\nmachine learning algorithms. Through rigorous experimental evaluations, this\npaper reviews the performance of machine learning approaches to posture\nrecognition, and analyses the robustness of the algorithms. When performing HAR\nwith the RGB-Depth data from our new dataset, machine learning algorithms such\nas a deep neural network reached a mean accuracy of up to 96.8% for\nclassification across all stationary and dynamic activities","url_abs":"http://arxiv.org/abs/1901.02858v1","url_pdf":"http://arxiv.org/pdf/1901.02858v1.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":"activity-recognition","task_name":"Activity Recognition"},{"task_slug":"autonomous-vehicles","task_name":"Autonomous Vehicles"},{"task_slug":"machine-learning","task_name":"BIG-bench Machine Learning"},{"task_slug":"human-activity-recognition","task_name":"Human Activity Recognition"},{"task_slug":"multimodal-activity-recognition","task_name":"Multimodal Activity Recognition"},{"task_slug":"sports-analytics","task_name":"Sports Analytics"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/multimodal-activity-recognition-on-lborohar","task":"Multimodal Activity Recognition","dataset":"LboroHAR","model":"Cubic SVM","rank_in_archive_order":1,"of":3,"metrics":{"Accuracy":"97.90"},"uses_additional_data":false},{"leaderboard":"/sota/multimodal-activity-recognition-on-lborohar","task":"Multimodal Activity Recognition","dataset":"LboroHAR","model":"Deep Neural Net","rank_in_archive_order":2,"of":3,"metrics":{"Accuracy":"95.00"},"uses_additional_data":false},{"leaderboard":"/sota/multimodal-activity-recognition-on-lborohar","task":"Multimodal Activity Recognition","dataset":"LboroHAR","model":"Bagged Trees","rank_in_archive_order":3,"of":3,"metrics":{"Accuracy":"92.50"},"uses_additional_data":false}],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}