{"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/centralnet-a-multilayer-approach-for","title":"CentralNet: a Multilayer Approach for Multimodal Fusion","arxiv_id":"1808.07275","date":"2018-08-22","proceeding":null,"authors":["Valentin Vielzeuf","Alexis Lechervy","Stéphane Pateux","Frédéric Jurie"],"abstract":"This paper proposes a novel multimodal fusion approach, aiming to produce\nbest possible decisions by integrating information coming from multiple media.\nWhile most of the past multimodal approaches either work by projecting the\nfeatures of different modalities into the same space, or by coordinating the\nrepresentations of each modality through the use of constraints, our approach\nborrows from both visions. More specifically, assuming each modality can be\nprocessed by a separated deep convolutional network, allowing to take decisions\nindependently from each modality, we introduce a central network linking the\nmodality specific networks. This central network not only provides a common\nfeature embedding but also regularizes the modality specific networks through\nthe use of multi-task learning. The proposed approach is validated on 4\ndifferent computer vision tasks on which it consistently improves the accuracy\nof existing multimodal fusion approaches.","url_abs":"http://arxiv.org/abs/1808.07275v1","url_pdf":"http://arxiv.org/pdf/1808.07275v1.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":"centralnet-a-multilayer-approach-for","repo_url":"https://github.com/jhaprince/multibully","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"centralnet-a-multilayer-approach-for","repo_url":"https://github.com/mengmenm/SMIL","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"multi-task-learning","task_name":"Multi-Task Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1808.07275","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}