{"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/middle-fusion-and-multi-stage-multi-form","title":"Middle Fusion and Multi-Stage, Multi-Form Prompts for Robust RGB-T Tracking","arxiv_id":"2403.18193","date":"2024-03-27","proceeding":null,"authors":["Qiming Wang","Yongqiang Bai","Hongxing Song"],"abstract":"RGB-T tracking, a vital downstream task of object tracking, has made remarkable progress in recent years. Yet, it remains hindered by two major challenges: 1) the trade-off between performance and efficiency; 2) the scarcity of training data. To address the latter challenge, some recent methods employ prompts to fine-tune pre-trained RGB tracking models and leverage upstream knowledge in a parameter-efficient manner. However, these methods inadequately explore modality-independent patterns and disregard the dynamic reliability of different modalities in open scenarios. We propose M3PT, a novel RGB-T prompt tracking method that leverages middle fusion and multi-modal and multi-stage visual prompts to overcome these challenges. We pioneer the use of the adjustable middle fusion meta-framework for RGB-T tracking, which could help the tracker balance the performance with efficiency, to meet various demands of application. Furthermore, based on the meta-framework, we utilize multiple flexible prompt strategies to adapt the pre-trained model to comprehensive exploration of uni-modal patterns and improved modeling of fusion-modal features in diverse modality-priority scenarios, harnessing the potential of prompt learning in RGB-T tracking. Evaluating on 6 existing challenging benchmarks, our method surpasses previous state-of-the-art prompt fine-tuning methods while maintaining great competitiveness against excellent full-parameter fine-tuning methods, with only 0.34M fine-tuned parameters.","url_abs":"https://arxiv.org/abs/2403.18193v2","url_pdf":"https://arxiv.org/pdf/2403.18193v2.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":"form","task_name":"Form"},{"task_slug":"object-tracking","task_name":"Object Tracking"},{"task_slug":"prompt-learning","task_name":"Prompt Learning"},{"task_slug":"rgb-t-tracking","task_name":"Rgb-T Tracking"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/rgb-t-tracking-on-lasher","task":"Rgb-T Tracking","dataset":"LasHeR","model":"M3PT","rank_in_archive_order":28,"of":39,"metrics":{"Precision":"67.3","Success":"54.2"},"uses_additional_data":false},{"leaderboard":"/sota/rgb-t-tracking-on-rgbt210","task":"Rgb-T Tracking","dataset":"RGBT210","model":"M3PT","rank_in_archive_order":14,"of":19,"metrics":{"Precision":"83.9","Success":"60.8"},"uses_additional_data":false},{"leaderboard":"/sota/rgb-t-tracking-on-rgbt234","task":"Rgb-T Tracking","dataset":"RGBT234","model":"M3PT","rank_in_archive_order":28,"of":42,"metrics":{"Precision":"85.9","Success":"63.4"},"uses_additional_data":false}],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}