{"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/modular-sensor-fusion-for-semantic","title":"Modular Sensor Fusion for Semantic Segmentation","arxiv_id":"1807.11249","date":"2018-07-30","proceeding":null,"authors":["Hermann Blum","Abel Gawel","Roland Siegwart","Cesar Cadena"],"abstract":"Sensor fusion is a fundamental process in robotic systems as it extends the\nperceptual range and increases robustness in real-world operations. Current\nmulti-sensor deep learning based semantic segmentation approaches do not\nprovide robustness to under-performing classes in one modality, or require a\nspecific architecture with access to the full aligned multi-sensor training\ndata. In this work, we analyze statistical fusion approaches for semantic\nsegmentation that overcome these drawbacks while keeping a competitive\nperformance. The studied approaches are modular by construction, allowing to\nhave different training sets per modality and only a much smaller subset is\nneeded to calibrate the statistical models. We evaluate a range of statistical\nfusion approaches and report their performance against state-of-the-art\nbaselines on both real-world and simulated data. In our experiments, the\napproach improves performance in IoU over the best single modality segmentation\nresults by up to 5%. We make all implementations and configurations publicly\navailable.","url_abs":"http://arxiv.org/abs/1807.11249v1","url_pdf":"http://arxiv.org/pdf/1807.11249v1.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":"modular-sensor-fusion-for-semantic","repo_url":"https://github.com/ethz-asl/modular_semantic_segmentation","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"segmentation","task_name":"Segmentation"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"},{"task_slug":"sensor-fusion","task_name":"Sensor Fusion"}],"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}