[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-85549-en":3,"doc-seo-85549-105":29,"detail-sidebar-cat-0-en-105":90},{"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":4,"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},85549,16904993612988,"Olivia Brown","https://ap-avatar.wpscdn.com/davatar_a8503ba1806abce46bf441b54a3ca4cd",8,"Research & Report","IP-SAM Rethinking Prompt-Conditioned Segmentation for Prompt-Absent Deployment","Prompt-conditioned foundation segmenters rely on explicit spatial prompts to guide mask decoding, but fully automatic deployment removes those prompts at inference, creating a structural mismatch. Existing feature-space adaptations often bypass the native prompt interface and weaken prompt-conditioned decoding. IP-SAM addresses this with prompt-space conditioning: a Self-Prompt Generator distills image context into intrinsic prompt anchors via SAM2’s frozen prompt encoder. Prompt-Space Gating uses intrinsic background prompts for asymmetric suppression. The method achieves state-of-the-art results on four camouflaged object detection benchmarks with only 21.26M trainable parameters and transfers to medical polyp segmentation.","arXiv :2603 .27250v2 [ cs .CV] 11 Jul 2026  \nIP-SAM: Rethinking Prompt-Conditioned Segmentation for Prompt-Absent Deployment  \nHuiyao Zhang 1 ,2 , Jin Bai 1 ,2 , Rui Guo 1 ,2 , JianWen Tan 1 ,2 , HongFei Wang  \n1 ,2 , and Ye Li 2 (􀀀)  \n1 University of Chinese Academy of Sciences,  \n2 Technology and Engineering Center for Space Utilization, Chinese Academy of Sciences  \n[zhanghuiyao25@csu.ac.cn](zhanghuiyao25@csu.ac.cn) , [baijin25@mails.ucas.ac.cn](baijin25@mails.ucas.ac.cn) , [liye@csu.ac.cn](liye@csu.ac.cn)  \nAbstract Prompt-conditioned foundation segmenters have emerged as a dominant paradigm for image segmentation, where explicit spatial prompts (e.g. , points, boxes, masks) guide mask decoding. However, many real-world deployments require fully automatic segmentation, creating a structural mismatch: the decoder expects prompts that are unavailable at inference. Existing adaptations typically modify intermediate features, inadvertently bypassing the model’s native prompt interface and weakening prompt-conditioned decoding. We propose IP-SAM, which revisits adaptation from a prompt-space perspective through prompt-space conditioning. Specifically, a Self-Prompt Generator (SPG) distills image context into complementary intrinsic prompts that serve as coarse regional anchors. These cues are projected through SAM2’s frozen prompt encoder, restoring prompt-guided decoding without external intervention. To suppress background-induced false positives, Prompt-Space Gating (PSG) leverages the intrinsic background prompt as an asymmetric suppressive constraint prior to decoding. Under a deterministic no-external-prompt protocol, IP-SAM achieves state-of-the-art performance across four camouflaged object detection benchmarks with only 21.26M trainable parameters. Furthermore, the proposed conditioning strategy generalizes beyond COD to medical polyp segmentation.  \nKeywords: Prompt-conditioned segmentation; Segment Anything Model; Camouflaged object detection; Parameter-efficient adaptation.  \n1 Introduction  \nPrompt-conditioned segmentation has emerged as a new paradigm for visual foundation models. Instead of directly predicting masks, these models rely on explicit spatial prompts—such as points, boxes, or masks—to guide the decoding process. Representative systems such as the Segment Anything Model (SAM) and its successor SAM2 [17, 35] demonstrate remarkable flexibility under interactive settings, enabling users to steer segmentation through simple spatial cues.  \nHowever, this paradigm reveals a fundamental mismatch in fully automatic deployment. In many real-world scenarios, segmentation must operate without  \n􀀀 Corresponding authors.  \n2 H. Zhang et al.  \nhuman interaction, leaving the model strictly prompt-absent at inference time. This creates a structural paradox: the decoder is designed to rely on promptconditioned signals, yet no prompts are available during deployment.  \nThis issue becomes particularly evident in Camouflaged Object Detection (COD), where targets blend into surrounding textures [6] . Without disambiguating prompts, models frequently activate visually similar background regions, leading to unstable masks and severe background leakage. Although we evaluate on COD as a challenging stress-test benchmark, the prompt-absent deployment problem arises broadly across automatic segmentation tasks.  \nTo address this challenge, most existing adaptations modify intermediate features by introducing adapters, lateral fusion modules, or task-specific decoders [4, 7] . While these feature-space modifications improve task alignment, they inadvertently bypass the native prompt interface that the architecture was originally designed to exploit. As a result, the prompt-conditioned decoding pathway remains weakly activated, leading to suboptimal adaptation. We argue that bypassing the prompt interface fundamentally breaks the design assumption of prompt-conditioned segmenters.  \nTo resolve this mismatch, we propose Intrinsic Pr","cbCaike05hHYbzFa","https://ap.wps.com/l/cbCaike05hHYbzFa","pdf",6133728,1,28,"English","en",105,"# Introduction\n## Prompt-conditioned segmentation and deployment mismatch\n## Existing adaptations and their limitations\n## Proposed IP-SAM: prompt-space conditioning\n## Evaluation setting and results","[{\"question\":\"Why does prompt-absent deployment create problems for prompt-conditioned segmenters?\",\"answer\":\"The decoder is designed to rely on prompt-guided signals, but inference provides no prompts in fully automatic deployment. This structural mismatch weakens segmentation quality.\"},{\"question\":\"How does IP-SAM generate prompts without external input?\",\"answer\":\"IP-SAM uses a Self-Prompt Generator to distill image context into intrinsic prompt anchors. These cues are projected through SAM2’s frozen prompt encoder to restore native prompt-guided decoding.\"},{\"question\":\"What is the role of Prompt-Space Gating (PSG) in IP-SAM?\",\"answer\":\"PSG leverages intrinsic background prompts as an asymmetric suppressive constraint before decoding. It filters deceptive activations to reduce background-induced false positives.\"}]",1784204461,71,{"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":85,"head_meta":87,"extra_data":89,"updated_unix":27},"ip-sam-rethinking-prompt-conditioned-segmentation-for-prompt-absent-deployment","",{"@graph":35,"@context":84},[36,53,67],{"@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/research-report/",3,{"item":51,"name":13,"@type":42,"position":52},"https://docshare.wps.com/document/ip-sam-rethinking-prompt-conditioned-segmentation-for-prompt-absent-deployment/85549/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":23,"description":14,"dateModified":61,"datePublished":61,"encodingFormat":60,"isAccessibleForFree":62,"interactionStatistic":63},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":40,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-16",true,{"@type":64,"interactionType":65,"userInteractionCount":4},"InteractionCounter",{"@type":66},"ViewAction",{"@type":68,"mainEntity":69},"FAQPage",[70,76,80],{"name":71,"@type":72,"acceptedAnswer":73},"Why does prompt-absent deployment create problems for prompt-conditioned segmenters?","Question",{"text":74,"@type":75},"The decoder is designed to rely on prompt-guided signals, but inference provides no prompts in fully automatic deployment. This structural mismatch weakens segmentation quality.","Answer",{"name":77,"@type":72,"acceptedAnswer":78},"How does IP-SAM generate prompts without external input?",{"text":79,"@type":75},"IP-SAM uses a Self-Prompt Generator to distill image context into intrinsic prompt anchors. These cues are projected through SAM2’s frozen prompt encoder to restore native prompt-guided decoding.",{"name":81,"@type":72,"acceptedAnswer":82},"What is the role of Prompt-Space Gating (PSG) in IP-SAM?",{"text":83,"@type":75},"PSG leverages intrinsic background prompts as an asymmetric suppressive constraint before decoding. It filters deceptive activations to reduce background-induced false positives.","https://schema.org",{"og:url":51,"og:type":86,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":88,"canonical":51},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":91},[92,96,100,104,109,114,119,122,127,130,134],{"id":20,"doc_module":4,"doc_module_name":45,"category_name":93,"show_sort_weight":94,"slug":95},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":97,"show_sort_weight":98,"slug":99},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":45,"category_name":101,"show_sort_weight":102,"slug":103},"Exam",70,"exam",{"id":105,"doc_module":4,"doc_module_name":45,"category_name":106,"show_sort_weight":107,"slug":108},5,"Comic",60,"comic",{"id":110,"doc_module":4,"doc_module_name":45,"category_name":111,"show_sort_weight":112,"slug":113},6,"Technology",50,"technology",{"id":115,"doc_module":4,"doc_module_name":45,"category_name":116,"show_sort_weight":117,"slug":118},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":120,"slug":121},30,"research-report",{"id":123,"doc_module":4,"doc_module_name":45,"category_name":124,"show_sort_weight":125,"slug":126},9,"Religion & Spirituality",20,"religion-spirituality",{"id":125,"doc_module":4,"doc_module_name":45,"category_name":128,"show_sort_weight":125,"slug":129},"World Cup","world-cup",{"id":131,"doc_module":4,"doc_module_name":45,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":45,"category_name":136,"show_sort_weight":105,"slug":137},19,"General","general"]