{"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/linear-support-tensor-machine-pedestrian","title":"Linear Support Tensor Machine: Pedestrian Detection in Thermal Infrared Images","arxiv_id":"1609.07878","date":"2016-09-26","proceeding":null,"authors":["Sujoy Kumar Biswas","Peyman Milanfar"],"abstract":"Pedestrian detection in thermal infrared images poses unique challenges\nbecause of the low resolution and noisy nature of the image. Here we propose a\nmid-level attribute in the form of multidimensional template, or tensor, using\nLocal Steering Kernel (LSK) as low-level descriptors for detecting pedestrians\nin far infrared images. LSK is specifically designed to deal with intrinsic\nimage noise and pixel level uncertainty by capturing local image geometry\nsuccinctly instead of collecting local orientation statistics (e.g., histograms\nin HOG). Our second contribution is the introduction of a new image similarity\nkernel in the popular maximum margin framework of support vector machines that\nresults in a relatively short and simple training phase for building a rigid\npedestrian detector. Our third contribution is to replace the sluggish but de\nfacto sliding window based detection methodology with multichannel discrete\nFourier transform, facilitating very fast and efficient pedestrian\nlocalization. The experimental studies on publicly available thermal infrared\nimages justify our proposals and model assumptions. In addition, the proposed\nwork also involves the release of our in-house annotations of pedestrians in\nmore than 17000 frames of OSU Color Thermal database for the purpose of sharing\nwith the research community.","url_abs":"http://arxiv.org/abs/1609.07878v1","url_pdf":"http://arxiv.org/pdf/1609.07878v1.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":"linear-support-tensor-machine-pedestrian","repo_url":"https://github.com/tigereatsheep/LSKfeatures","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"attribute","task_name":"Attribute"},{"task_slug":"pedestrian-detection","task_name":"Pedestrian Detection"}],"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}