[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-83410-en":3,"doc-seo-83410-105":29,"detail-sidebar-cat-0-en-105":83},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":13,"seo_description":14,"update_tm":27,"read_time":28},83410,7971461741311,"Ophelia","https://ap-avatar.wpscdn.com/avatar/74000253aff267980c6?x-image-process=image/resize,m_fixed,w_180,h_180&k=1779345379180704826",6,"Technology","SAM-MT: Real-Time Interactive Multi-Target Video Segmentation","SAM-MT is a real-time interactive framework for multi-target video object segmentation built upon Segment Anything 2 (SAM2). It converts the single-target workflow into an interactive setting using explicit target queries in parallel with a shared representation for global context. Decoupled masked attention limits cross-target interference, while query-based sparse memory supports stable temporal evolution and per-target re-identification. Specialized occlusion handling and overlap prevention enable near-single-object efficiency, achieving 36+ FPS with 10 targets while matching SAM2 performance.","arXiv :2607 .08688v 1 [ cs .CV] 9 Jul 2026  \nSAM-MT: Real-Time Interactive Multi-Target Video Segmentation Ruiqi Shen 1 , Chang Liu2 , Henghui Ding 1  \n1 Fudan University, 2 Shanghai University of Finance and Economics  \nModern Video Object Segmentation (VOS) involves tracking and segmenting user-specified targets. While recent approaches have achieved remarkable performance in single-target scenarios, extending them to multi-target settings typically involves replicating the single-target processing for each individual object, resulting in reduced frame rates (FPS) with unbounded latency as target count increases. Built upon Segment Anything 2 (SAM2), we propose SAM-MT, which addresses this by transforming the model into an interactive framework for real-time Multi-Target video segmentation. SAM-MT uses explicit queries to represent different individual targets, in parallel with a shared representation for global context. It employs decoupled masked attention to keep individual identities distinct from cross-target interference, and sparse memory for stable temporal evolution, along with specialized strategies for occlusion handling and overlap prevention. SAM-MT successfully decouples latency from the number of targets, achieving real-time speed on par with single-target baselines (>36 FPS for 10 targets) while maintaining SAM2’s robust video segmentation performance.  \nCode: [https://github.com/FudanCVL/SAM-MT](https://github.com/FudanCVL/SAM-MT)  \nEmail: [henghui.ding@gmail.com](henghui.ding@gmail.com)  \n1 Introduction  \nContemporary Video Object Segmentation (VOS) [1, 2] tracks and segments user-specified objects in openvocabulary environments, with applications spanning in-the-wild navigation [3] and robotics [4] . Dominant approaches follow the space-time-memory (STM) paradigm [5], exploiting heavy pixel-level representations stored in a dense memory to track and segment target and have achieved state-of-the-art (SOTA) performance across VOS benchmarks [5–12] .  \nWhile highly impressive, most of these approaches, including the powerful SAM2 [9] and its extensions [10, 13 , 14], rely on object-wise memory and propagation, making them tailored for single-target processing. Therefore, handling multiple targets requires repeating the same computations for every additional instance, as illustrated in Figure 1(a) . Even with some earlier attempts to share frame-level image features, they still rely on independent object-wise processing for mask decoding and memory encoding [6–8] (Figure 1(b)) . Consequently, such a straightforward extension results in a substantial rise in computational cost as the number of targets increases, leading to degraded frame rates (FPS) with unbounded latency. One possible workaround is to merge all targets into a single trackable object. However, this approach discards individual target IDs, and is therefore impractical for real-world applications. Furthermore, the merged object often forms irregular shapes that standard VOS models, trained on real-world objects, simply fail to recognize and track (see Section 5.3) .  \nMeanwhile, real-world applications demand real-time tracking and segmentation of multiple targets. For instance, a practical VOS system for autonomous driving must 1) preserve individual target identities during propagation while 2) maintaining real-time speed, even in crowded scenes with dense vehicles and pedestrians. These requirements highlight the need for a real-time multi-target video segmentation framework that could maintain near-single-object efficiency as target count increases, while faithfully preserving and updating individual identities during propagation. Recent approaches, including SAM2, struggle to bridge this gap, as their object-wise processing causes computation to grow with target count.  \nWe therefore present SAM-MT to address this. As illustrated in Figure 1(c), SAM-MT extends the SAM2 architecture with a hybrid approach: it employs a shared representation t","cbCair6OJ3kkidwD","https://ap.wps.com/l/cbCair6OJ3kkidwD","pdf",9729130,1,14,"English","en",105,"# Introduction\n## Problem with existing single-target extensions\n## Motivation from real-world real-time needs\n## Proposed SAM-MT approach","[{\"question\":\"What enables SAM-MT to achieve real-time performance as the number of targets increases?\",\"answer\":\"It uses a shared representation for global context and lightweight target queries, avoiding redundant object-wise mask decoding and memory encoding. A query-based sparse memory supports stable temporal evolution without extra heavy components.\"}]",1784187391,35,{"code":4,"msg":30,"data":31},"ok",{"site_id":24,"language":23,"slug":32,"title":13,"keywords":33,"description":14,"schema_data":34,"social_meta":78,"head_meta":80,"extra_data":82,"updated_unix":27},"sam-mt-real-time-interactive-multi-target-video-segmentation","",{"@graph":35,"@context":77},[36,53,68],{"@type":37,"itemListElement":38},"BreadcrumbList",[39,43,47,50],{"item":40,"name":41,"@type":42,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":44,"name":45,"@type":42,"position":46},"https://docshare.wps.com/document/","Document",2,{"item":48,"name":12,"@type":42,"position":49},"https://docshare.wps.com/document/technology/",3,{"item":51,"name":13,"@type":42,"position":52},"https://docshare.wps.com/document/sam-mt-real-time-interactive-multi-target-video-segmentation/83410/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":23,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":40,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-17","2026-07-16",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71],{"name":72,"@type":73,"acceptedAnswer":74},"What enables SAM-MT to achieve real-time performance as the number of targets increases?","Question",{"text":75,"@type":76},"It uses a shared representation for global context and lightweight target queries, avoiding redundant object-wise mask decoding and memory encoding. A query-based sparse memory supports stable temporal evolution without extra heavy components.","Answer","https://schema.org",{"og:url":51,"og:type":79,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":81,"canonical":51},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":84},[85,89,93,97,102,105,110,115,120,123,127],{"id":20,"doc_module":4,"doc_module_name":45,"category_name":86,"show_sort_weight":87,"slug":88},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":90,"show_sort_weight":91,"slug":92},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":45,"category_name":94,"show_sort_weight":95,"slug":96},"Exam",70,"exam",{"id":98,"doc_module":4,"doc_module_name":45,"category_name":99,"show_sort_weight":100,"slug":101},5,"Comic",60,"comic",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":103,"slug":104},50,"technology",{"id":106,"doc_module":4,"doc_module_name":45,"category_name":107,"show_sort_weight":108,"slug":109},7,"Healthcare",40,"healthcare",{"id":111,"doc_module":4,"doc_module_name":45,"category_name":112,"show_sort_weight":113,"slug":114},8,"Research & Report",30,"research-report",{"id":116,"doc_module":4,"doc_module_name":45,"category_name":117,"show_sort_weight":118,"slug":119},9,"Religion & Spirituality",20,"religion-spirituality",{"id":118,"doc_module":4,"doc_module_name":45,"category_name":121,"show_sort_weight":118,"slug":122},"World Cup","world-cup",{"id":124,"doc_module":4,"doc_module_name":45,"category_name":125,"show_sort_weight":124,"slug":126},10,"Lifestyle","lifestyle",{"id":128,"doc_module":4,"doc_module_name":45,"category_name":129,"show_sort_weight":98,"slug":130},19,"General","general"]