{"id":"https://openalex.org/W4417469609","doi":"https://doi.org/10.48550/arxiv.2512.14440","title":"S2D: Sparse-To-Dense Keymask Distillation for Unsupervised Video Instance Segmentation","display_name":"S2D: Sparse-To-Dense Keymask Distillation for Unsupervised Video Instance Segmentation","publication_year":2025,"publication_date":"2025-12-16","ids":{"openalex":"https://openalex.org/W4417469609","doi":"https://doi.org/10.48550/arxiv.2512.14440"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2512.14440","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2512.14440","pdf_url":"https://arxiv.org/pdf/2512.14440","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2512.14440","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5092772246","display_name":"Leon Sick","orcid":"https://orcid.org/0009-0004-6524-0715","observed_orcids":["https://orcid.org/0009-0004-6524-0715"]},"institutions":[{"id":"https://openalex.org/I196349391","display_name":"Universit\u00e4t Ulm","ror":"https://ror.org/032000t02","country_code":"DE","type":"education","lineage":["https://openalex.org/I196349391"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Sick, Leon","raw_affiliation_strings":["Ulm University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Ulm University","institution_ids":["https://openalex.org/I196349391"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5053328232","display_name":"Lukas Hoyer","orcid":"https://orcid.org/0000-0002-7391-0676","observed_orcids":["https://orcid.org/0000-0002-7391-0676"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hoyer, Lukas","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5066189200","display_name":"Dominik Engel","orcid":"https://orcid.org/0000-0002-5766-7215","observed_orcids":["https://orcid.org/0000-0002-5766-7215"]},"institutions":[{"id":"https://openalex.org/I145847075","display_name":"TU Wien","ror":"https://ror.org/04d836q62","country_code":"AT","type":"education","lineage":["https://openalex.org/I145847075"]}],"countries":["AT"],"is_corresponding":false,"raw_author_name":"Engel, Dominik","raw_affiliation_strings":["TU Vienna"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"TU Vienna","institution_ids":["https://openalex.org/I145847075"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5088394147","display_name":"Pedro Hermosilla","orcid":"https://orcid.org/0000-0003-3586-4741","observed_orcids":["https://orcid.org/0000-0003-3586-4741"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hermosilla, Pedro","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5061059879","display_name":"Timo Ropinski","orcid":"https://orcid.org/0000-0002-7857-5512","observed_orcids":["https://orcid.org/0000-0002-7857-5512"]},"institutions":[{"id":"https://openalex.org/I196349391","display_name":"Universit\u00e4t Ulm","ror":"https://ror.org/032000t02","country_code":"DE","type":"education","lineage":["https://openalex.org/I196349391"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Ropinski, Timo","raw_affiliation_strings":["Ulm University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Ulm University","institution_ids":["https://openalex.org/I196349391"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.8407999873161316,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.8407999873161316,"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/T10331","display_name":"Video Surveillance and Tracking Methods","score":0.03720000013709068,"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/T10052","display_name":"Medical Image Segmentation Techniques","score":0.029500000178813934,"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/pseudo-labeling","display_name":"pseudo-labeling","score":0.6620000004768372},{"id":"https://openalex.org/keywords/temporal-coherence","display_name":"temporal coherence","score":0.8289999961853027},{"id":"https://openalex.org/keywords/temporal-noise","display_name":"temporal noise","score":0.5979999899864197}],"concepts":[{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.8237000107765198},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7946000099182129},{"id":"https://openalex.org/C89600930","wikidata":"https://www.wikidata.org/wiki/Q1423946","display_name":"Segmentation","level":2,"score":0.7071999907493591},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.6435999870300293},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.4975000023841858},{"id":"https://openalex.org/C124504099","wikidata":"https://www.wikidata.org/wiki/Q56933","display_name":"Image segmentation","level":3,"score":0.45320001244544983},{"id":"https://openalex.org/C65885262","wikidata":"https://www.wikidata.org/wiki/Q7429708","display_name":"Scale-space segmentation","level":4,"score":0.4291999936103821},{"id":"https://openalex.org/C12713177","wikidata":"https://www.wikidata.org/wiki/Q1900281","display_name":"Perspective (graphical)","level":2,"score":0.3955000042915344},{"id":"https://openalex.org/C30814859","wikidata":"https://www.wikidata.org/wiki/Q4119603","display_name":"Video denoising","level":5,"score":0.3797999918460846},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.36980000138282776},{"id":"https://openalex.org/C8038995","wikidata":"https://www.wikidata.org/wiki/Q1152135","display_name":"Unsupervised learning","level":2,"score":0.367000013589859},{"id":"https://openalex.org/C128840427","wikidata":"https://www.wikidata.org/wiki/Q1302174","display_name":"Motion compensation","level":2,"score":0.36649999022483826},{"id":"https://openalex.org/C25694479","wikidata":"https://www.wikidata.org/wiki/Q7446278","display_name":"Segmentation-based object categorization","level":5,"score":0.3573000133037567},{"id":"https://openalex.org/C2781181686","wikidata":"https://www.wikidata.org/wiki/Q4226068","display_name":"Coherence (philosophical gambling strategy)","level":2,"score":0.33320000767707825},{"id":"https://openalex.org/C104114177","wikidata":"https://www.wikidata.org/wiki/Q79782","display_name":"Motion (physics)","level":2,"score":0.32710000872612},{"id":"https://openalex.org/C10161872","wikidata":"https://www.wikidata.org/wiki/Q557891","display_name":"Motion estimation","level":2,"score":0.3215000033378601},{"id":"https://openalex.org/C75553542","wikidata":"https://www.wikidata.org/wiki/Q178161","display_name":"A priori and a posteriori","level":2,"score":0.320499986410141},{"id":"https://openalex.org/C160920958","wikidata":"https://www.wikidata.org/wiki/Q7662746","display_name":"Synthetic data","level":2,"score":0.31459999084472656},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.2770000100135803},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.2662999927997589},{"id":"https://openalex.org/C202474056","wikidata":"https://www.wikidata.org/wiki/Q1931635","display_name":"Video tracking","level":3,"score":0.26489999890327454},{"id":"https://openalex.org/C52622490","wikidata":"https://www.wikidata.org/wiki/Q1026626","display_name":"Feature extraction","level":2,"score":0.25870001316070557}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2512.14440","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2512.14440","pdf_url":"https://arxiv.org/pdf/2512.14440","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2512.14440","doi":"https://doi.org/10.48550/arxiv.2512.14440","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2512.14440","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2512.14440","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2512.14440","pdf_url":"https://arxiv.org/pdf/2512.14440","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4417469609.pdf","grobid_xml":"https://content.openalex.org/works/W4417469609.grobid-xml"},"referenced_works_count":3,"referenced_works":["https://openalex.org/W4282941252","https://openalex.org/W4401307635","https://openalex.org/W6979267127"],"related_works":["https://openalex.org/W4417287186","https://openalex.org/W4402727608","https://openalex.org/W4402917031","https://openalex.org/W4404204177","https://openalex.org/W7113915225","https://openalex.org/W2949735689","https://openalex.org/W7078324429","https://openalex.org/W4416408606","https://openalex.org/W4400222601","https://openalex.org/W3212561242"],"abstract_inverted_index":{"In":[0],"recent":[1],"years,":[2],"the":[3,100,112,150,154,161],"state-of-the-art":[4,163],"in":[5,42,111],"unsupervised":[6,66,80],"video":[7,15,26,67,75,86,113],"instance":[8,34,68,81],"segmentation":[9,69,82,129],"has":[10],"heavily":[11],"relied":[12],"on":[13,73,84,153],"synthetic":[14],"data,":[16],"generated":[17],"from":[18,79],"object-centric":[19],"image":[20,33],"datasets":[21],"such":[22,44],"as":[23,45],"ImageNet.":[24],"However,":[25,88],"synthesis":[27],"by":[28,49,107,114,144],"artificially":[29],"shifting":[30],"and":[31,95],"scaling":[32],"masks":[35,83],"fails":[36],"to":[37,126],"accurately":[38],"model":[39,70,130,152],"realistic":[40],"motion":[41,117],"videos,":[43],"perspective":[46],"changes,":[47],"movement":[48],"parts":[50],"of":[51],"one":[52],"or":[53,56],"multiple":[54],"instances,":[55],"camera":[57],"motion.":[58],"To":[59],"tackle":[60],"this":[61],"issue,":[62],"we":[63,103,137],"propose":[64,138],"an":[65],"trained":[71],"exclusively":[72],"real":[74],"data.":[76],"We":[77],"start":[78],"individual":[85],"frames.":[87],"these":[89],"single-frame":[90],"segmentations":[91],"exhibit":[92],"temporal":[93,105],"noise":[94],"their":[96],"quality":[97],"varies":[98],"through":[99],"video.":[101],"Therefore,":[102],"establish":[104],"coherence":[106],"identifying":[108],"high-quality":[109],"keymasks":[110],"leveraging":[115],"deep":[116],"priors.":[118],"The":[119],"sparse":[120],"keymask":[121],"pseudo-annotations":[122],"are":[123],"then":[124],"used":[125],"train":[127],"a":[128,139,145],"for":[131,135],"implicit":[132],"mask":[133],"propagation,":[134],"which":[136],"Sparse-To-Dense":[140],"Distillation":[141],"approach":[142,159],"aided":[143],"Temporal":[146],"DropLoss.":[147],"After":[148],"training":[149],"final":[151],"resulting":[155],"dense":[156],"labelset,":[157],"our":[158],"outperforms":[160],"current":[162],"across":[164],"various":[165],"benchmarks.":[166]},"counts_by_year":[],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2025-12-18T00:00:00"}