{"id":"https://openalex.org/W7139117809","doi":"https://doi.org/10.48550/arxiv.2603.17813","title":"M2P: Improving Visual Foundation Models with Mask-to-Point Weakly-Supervised Learning for Dense Point Tracking","display_name":"M2P: Improving Visual Foundation Models with Mask-to-Point Weakly-Supervised Learning for Dense Point Tracking","publication_year":2026,"publication_date":"2026-03-18","ids":{"openalex":"https://openalex.org/W7139117809","doi":"https://doi.org/10.48550/arxiv.2603.17813"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2603.17813","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.17813","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2603.17813","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5062279616","display_name":"Qiangqiang Wu","orcid":"https://orcid.org/0000-0002-3847-7838"},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Wu, Qiangqiang","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129950300","display_name":"Tianyu Yang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yang, Tianyu","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129821463","display_name":"Bo Fang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Fang, Bo","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129879380","display_name":"Jia Wan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wan, Jia","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5127880208","display_name":"Matias Di Martino","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Di Martino, Matias","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130030160","display_name":"Guillermo Sapiro","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sapiro, Guillermo","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5130138159","display_name":"Antoni B. Chan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chan, Antoni B.","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":7,"corresponding_author_ids":["https://openalex.org/A5062279616"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10331","display_name":"Video Surveillance and Tracking Methods","score":0.6800000071525574,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10331","display_name":"Video Surveillance and Tracking Methods","score":0.6800000071525574,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10812","display_name":"Human Pose and Action Recognition","score":0.05220000073313713,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10531","display_name":"Advanced Vision and Imaging","score":0.03240000084042549,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/consistency","display_name":"Consistency (knowledge bases)","score":0.6714000105857849},{"id":"https://openalex.org/keywords/segmentation","display_name":"Segmentation","score":0.58160001039505},{"id":"https://openalex.org/keywords/matching","display_name":"Matching (statistics)","score":0.5194000005722046},{"id":"https://openalex.org/keywords/point","display_name":"Point (geometry)","score":0.5121999979019165},{"id":"https://openalex.org/keywords/representation","display_name":"Representation (politics)","score":0.5065000057220459},{"id":"https://openalex.org/keywords/video-tracking","display_name":"Video tracking","score":0.47510001063346863},{"id":"https://openalex.org/keywords/object","display_name":"Object (grammar)","score":0.46140000224113464},{"id":"https://openalex.org/keywords/code","display_name":"Code (set theory)","score":0.43880000710487366},{"id":"https://openalex.org/keywords/tracking","display_name":"Tracking (education)","score":0.4156000018119812}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7667999863624573},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6966999769210815},{"id":"https://openalex.org/C2776436953","wikidata":"https://www.wikidata.org/wiki/Q5163215","display_name":"Consistency (knowledge bases)","level":2,"score":0.6714000105857849},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.6061000227928162},{"id":"https://openalex.org/C89600930","wikidata":"https://www.wikidata.org/wiki/Q1423946","display_name":"Segmentation","level":2,"score":0.58160001039505},{"id":"https://openalex.org/C165064840","wikidata":"https://www.wikidata.org/wiki/Q1321061","display_name":"Matching (statistics)","level":2,"score":0.5194000005722046},{"id":"https://openalex.org/C28719098","wikidata":"https://www.wikidata.org/wiki/Q44946","display_name":"Point (geometry)","level":2,"score":0.5121999979019165},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.5065000057220459},{"id":"https://openalex.org/C202474056","wikidata":"https://www.wikidata.org/wiki/Q1931635","display_name":"Video tracking","level":3,"score":0.47510001063346863},{"id":"https://openalex.org/C2781238097","wikidata":"https://www.wikidata.org/wiki/Q175026","display_name":"Object (grammar)","level":2,"score":0.46140000224113464},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.43880000710487366},{"id":"https://openalex.org/C2775936607","wikidata":"https://www.wikidata.org/wiki/Q466845","display_name":"Tracking (education)","level":2,"score":0.4156000018119812},{"id":"https://openalex.org/C2777303404","wikidata":"https://www.wikidata.org/wiki/Q759757","display_name":"Convergence (economics)","level":2,"score":0.40709999203681946},{"id":"https://openalex.org/C62354387","wikidata":"https://www.wikidata.org/wiki/Q875399","display_name":"Boundary (topology)","level":2,"score":0.39570000767707825},{"id":"https://openalex.org/C155542232","wikidata":"https://www.wikidata.org/wiki/Q736111","display_name":"Optical flow","level":3,"score":0.34299999475479126},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.34290000796318054},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.34150001406669617},{"id":"https://openalex.org/C56461940","wikidata":"https://www.wikidata.org/wiki/Q970687","display_name":"Eye tracking","level":2,"score":0.3407999873161316},{"id":"https://openalex.org/C59404180","wikidata":"https://www.wikidata.org/wiki/Q17013334","display_name":"Feature learning","level":2,"score":0.34040001034736633},{"id":"https://openalex.org/C104114177","wikidata":"https://www.wikidata.org/wiki/Q79782","display_name":"Motion (physics)","level":2,"score":0.33160001039505005},{"id":"https://openalex.org/C12725497","wikidata":"https://www.wikidata.org/wiki/Q810247","display_name":"Baseline (sea)","level":2,"score":0.3303000032901764},{"id":"https://openalex.org/C124504099","wikidata":"https://www.wikidata.org/wiki/Q56933","display_name":"Image segmentation","level":3,"score":0.32659998536109924},{"id":"https://openalex.org/C2776151529","wikidata":"https://www.wikidata.org/wiki/Q3045304","display_name":"Object detection","level":3,"score":0.30230000615119934},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.28679999709129333},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.28439998626708984},{"id":"https://openalex.org/C10161872","wikidata":"https://www.wikidata.org/wiki/Q557891","display_name":"Motion estimation","level":2,"score":0.27390000224113464},{"id":"https://openalex.org/C179458375","wikidata":"https://www.wikidata.org/wiki/Q1020763","display_name":"Bundle adjustment","level":3,"score":0.2671000063419342},{"id":"https://openalex.org/C64876066","wikidata":"https://www.wikidata.org/wiki/Q5141226","display_name":"Cognitive neuroscience of visual object recognition","level":3,"score":0.26190000772476196},{"id":"https://openalex.org/C95020103","wikidata":"https://www.wikidata.org/wiki/Q1813492","display_name":"Match moving","level":3,"score":0.2554999887943268},{"id":"https://openalex.org/C83248878","wikidata":"https://www.wikidata.org/wiki/Q344000","display_name":"Active appearance model","level":3,"score":0.2515999972820282}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2603.17813","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.17813","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2603.17813","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.17813","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Tracking":[0],"Any":[1],"Point":[2],"(TAP)":[3],"has":[4],"emerged":[5],"as":[6,140,208,225],"a":[7,85,103,115,141],"fundamental":[8],"tool":[9],"for":[10,40,67,78,211,229],"video":[11,58],"understanding.":[12],"Current":[13],"approaches":[14],"adapt":[15],"Vision":[16],"Foundation":[17],"Models":[18],"(VFMs)":[19],"like":[20],"DINOv2":[21],"via":[22],"offline":[23,216],"finetuning":[24],"or":[25],"test-time":[26,213],"optimization.":[27],"However,":[28],"these":[29],"VFMs":[30,66,173],"rely":[31],"on":[32,196],"static":[33],"image":[34],"pre-training,":[35],"which":[36,55,90,121,143],"is":[37,156],"inherently":[38],"sub-optimal":[39],"capturing":[41],"dense":[42,68],"temporal":[43],"correspondence":[44],"in":[45],"videos.":[46,183],"To":[47],"address":[48],"this,":[49],"we":[50,83,113],"propose":[51,84,114],"Mask-to-Point":[52],"(M2P)":[53],"learning,":[54],"leverages":[56,91],"rich":[57],"object":[59],"segmentation":[60],"(VOS)":[61],"mask":[62,116,153],"annotations":[63],"to":[64,94,149,158,223],"improve":[65],"point":[69,230],"tracking.":[70,231],"Our":[71],"M2P":[72,168,185,204],"introduces":[73],"three":[74],"new":[75],"mask-based":[76],"constraints":[77],"weakly-supervised":[79],"representation":[80],"learning.":[81,111],"First,":[82],"local":[86,104],"structure":[87],"consistency":[88,118],"loss,":[89,120],"Procrustes":[92],"analysis":[93],"model":[95],"the":[96,197,202],"cohesive":[97],"motion":[98],"of":[99],"points":[100,126],"lying":[101],"within":[102],"structure,":[105],"achieving":[106],"more":[107],"reliable":[108],"point-to-point":[109],"matching":[110],"Second,":[112],"label":[117],"(MLC)":[119],"enforces":[122],"that":[123,165],"sampled":[124],"foreground":[125,129],"strictly":[127],"match":[128],"regions":[130],"across":[131],"frames.":[132],"The":[133],"proposed":[134,203],"MLC":[135],"loss":[136],"can":[137],"be":[138,234],"regarded":[139],"regularization,":[142],"stabilizes":[144],"training":[145,176,182],"and":[146,188,194,215],"prevents":[147],"convergence":[148],"trivial":[150],"solutions.":[151],"Finally,":[152],"boundary":[154,161],"constrain":[155],"applied":[157],"explicitly":[159],"supervise":[160],"points.":[162],"We":[163],"show":[164],"our":[166],"weaklysupervised":[167],"models":[169,205,228],"significantly":[170],"outperform":[171],"baseline":[172],"with":[174],"efficient":[175],"by":[177],"using":[178],"only":[179],"3.6K":[180],"VOS":[181],"Notably,":[184],"achieves":[186],"12.8%":[187],"14.6%":[189],"performance":[190],"gains":[191],"over":[192],"DINOv2-B/14":[193],"DINOv3-B/16":[195],"TAP-Vid-DAVIS":[198],"benchmark,":[199],"respectively.":[200],"Moreover,":[201],"are":[206],"used":[207],"pre-trained":[209,227],"backbones":[210],"both":[212],"optimized":[214],"fine-tuned":[217],"TAP":[218],"tasks,":[219],"demonstrating":[220],"its":[221],"potential":[222],"serve":[224],"general":[226],"Code":[232],"will":[233],"made":[235],"publicly":[236],"available":[237],"upon":[238],"acceptance.":[239]},"counts_by_year":[],"updated_date":"2026-03-20T20:54:20.808490","created_date":"2026-03-20T00:00:00"}