{"id":"https://openalex.org/W7225129488","doi":"https://doi.org/10.21203/rs.3.rs-10799029/v1","title":"RGPFormer: Input-Adaptive Geometric Bias Calibration for RGB-D Semantic Segmentation","display_name":"RGPFormer: Input-Adaptive Geometric Bias Calibration for RGB-D Semantic Segmentation","publication_year":2026,"publication_date":"2026-11-02","ids":{"openalex":"https://openalex.org/W7225129488","doi":"https://doi.org/10.21203/rs.3.rs-10799029/v1"},"language":"en","primary_location":{"id":"doi:10.21203/rs.3.rs-10799029/v1","doi":"https://doi.org/10.21203/rs.3.rs-10799029/v1","is_oa":true,"landing_page_url":"https://doi.org/10.21203/rs.3.rs-10799029/v1","pdf_url":"https://www.researchsquare.com/article/rs-10799029/latest.pdf","source":{"id":"https://openalex.org/S4306525896","display_name":"Research Square","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"acceptedVersion","is_accepted":true,"is_published":false,"raw_source_name":"Research Square","raw_type":"posted-content"},"volume":null,"type":"preprint","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://www.researchsquare.com/article/rs-10799029/latest.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5140226668","display_name":"Enhua Liu","orcid":"https://orcid.org/0009-0009-7298-0814","observed_orcids":["https://orcid.org/0009-0009-7298-0814"]},"institutions":[{"id":"https://openalex.org/I107982497","display_name":"Dalian Ocean University","ror":"https://ror.org/0523b6g79","country_code":"CN","type":"education","lineage":["https://openalex.org/I107982497"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Enhua Liu","raw_affiliation_strings":["Dalian Ocean University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Dalian Ocean University","institution_ids":["https://openalex.org/I107982497"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5160374014","display_name":"Zitong Li","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I107982497","display_name":"Dalian Ocean University","ror":"https://ror.org/0523b6g79","country_code":"CN","type":"education","lineage":["https://openalex.org/I107982497"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zitong Li","raw_affiliation_strings":["Dalian Ocean University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Dalian Ocean University","institution_ids":["https://openalex.org/I107982497"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5160356394","display_name":"Yi He","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I4405282062","display_name":"Shandong Transport Vocational College","ror":"https://ror.org/0591y1787","country_code":"CN","type":"education","lineage":["https://openalex.org/I4405282062"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yi He","raw_affiliation_strings":["Shandong Transport Vocational College"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Shandong Transport Vocational College","institution_ids":["https://openalex.org/I4405282062"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5140292223","display_name":"Chengwang Han","orcid":"https://orcid.org/0009-0004-8351-3158","observed_orcids":["https://orcid.org/0009-0004-8351-3158"]},"institutions":[{"id":"https://openalex.org/I107982497","display_name":"Dalian Ocean University","ror":"https://ror.org/0523b6g79","country_code":"CN","type":"education","lineage":["https://openalex.org/I107982497"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Chengwang Han","raw_affiliation_strings":["Dalian Ocean University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Dalian Ocean University","institution_ids":["https://openalex.org/I107982497"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5140235987","display_name":"Liqiang Chang","orcid":"https://orcid.org/0009-0000-5209-6399","observed_orcids":["https://orcid.org/0009-0000-5209-6399"]},"institutions":[{"id":"https://openalex.org/I107982497","display_name":"Dalian Ocean University","ror":"https://ror.org/0523b6g79","country_code":"CN","type":"education","lineage":["https://openalex.org/I107982497"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Liqiang Chang","raw_affiliation_strings":["Dalian Ocean University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Dalian Ocean University","institution_ids":["https://openalex.org/I107982497"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5160378089","display_name":"Xiang Li","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I107982497","display_name":"Dalian Ocean University","ror":"https://ror.org/0523b6g79","country_code":"CN","type":"education","lineage":["https://openalex.org/I107982497"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xiang Li","raw_affiliation_strings":["Dalian Ocean University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Dalian Ocean University","institution_ids":["https://openalex.org/I107982497"]}]}],"institutions":[],"countries_distinct_count":1,"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":0.9854000210762024,"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":0.9854000210762024,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.007600000128149986,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"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":0.0027000000700354576,"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/rgb-d-semantic-segmentation","display_name":"RGB-D semantic segmentation","score":0.996999979019165},{"id":"https://openalex.org/keywords/depth-bias","display_name":"depth bias","score":0.40400001406669617},{"id":"https://openalex.org/keywords/attention-mechanism","display_name":"attention mechanism","score":0.4659999907016754},{"id":"https://openalex.org/keywords/depth-gradient","display_name":"depth gradient","score":0.4959999918937683},{"id":"https://openalex.org/keywords/transformer-encoder","display_name":"Transformer encoder","score":0.5720000267028809}],"concepts":[{"id":"https://openalex.org/C99844830","wikidata":"https://www.wikidata.org/wiki/Q102441924","display_name":"Scaling","level":2,"score":0.6384999752044678},{"id":"https://openalex.org/C89600930","wikidata":"https://www.wikidata.org/wiki/Q1423946","display_name":"Segmentation","level":2,"score":0.5942000150680542},{"id":"https://openalex.org/C155512373","wikidata":"https://www.wikidata.org/wiki/Q287450","display_name":"Residual","level":2,"score":0.5605000257492065},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5579000115394592},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.555400013923645},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.4846999943256378},{"id":"https://openalex.org/C198082294","wikidata":"https://www.wikidata.org/wiki/Q3399648","display_name":"Position (finance)","level":2,"score":0.4706999957561493},{"id":"https://openalex.org/C184898388","wikidata":"https://www.wikidata.org/wiki/Q1435712","display_name":"Pairwise comparison","level":2,"score":0.4661000072956085},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.41830000281333923},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.39910000562667847},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.3659999966621399},{"id":"https://openalex.org/C52672216","wikidata":"https://www.wikidata.org/wiki/Q1749840","display_name":"Depth perception","level":3,"score":0.3422999978065491},{"id":"https://openalex.org/C28719098","wikidata":"https://www.wikidata.org/wiki/Q44946","display_name":"Point (geometry)","level":2,"score":0.34200000762939453},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.33559998869895935},{"id":"https://openalex.org/C64543145","wikidata":"https://www.wikidata.org/wiki/Q162942","display_name":"Intersection (aeronautics)","level":2,"score":0.3206000030040741},{"id":"https://openalex.org/C131979681","wikidata":"https://www.wikidata.org/wiki/Q1899648","display_name":"Point cloud","level":2,"score":0.31929999589920044},{"id":"https://openalex.org/C68537008","wikidata":"https://www.wikidata.org/wiki/Q247932","display_name":"Stereopsis","level":2,"score":0.31349998712539673},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.3118000030517578},{"id":"https://openalex.org/C65909025","wikidata":"https://www.wikidata.org/wiki/Q1945033","display_name":"Monocular","level":2,"score":0.31150001287460327},{"id":"https://openalex.org/C2776502983","wikidata":"https://www.wikidata.org/wiki/Q690182","display_name":"Contrast (vision)","level":2,"score":0.30720001459121704},{"id":"https://openalex.org/C165838908","wikidata":"https://www.wikidata.org/wiki/Q736777","display_name":"Calibration","level":2,"score":0.3034999966621399},{"id":"https://openalex.org/C43711488","wikidata":"https://www.wikidata.org/wiki/Q7534783","display_name":"Skew","level":2,"score":0.29600000381469727},{"id":"https://openalex.org/C141268832","wikidata":"https://www.wikidata.org/wiki/Q2940499","display_name":"Depth map","level":3,"score":0.2937999963760376},{"id":"https://openalex.org/C142259097","wikidata":"https://www.wikidata.org/wiki/Q5891314","display_name":"Homogeneity (statistics)","level":2,"score":0.29190000891685486},{"id":"https://openalex.org/C124504099","wikidata":"https://www.wikidata.org/wiki/Q56933","display_name":"Image segmentation","level":3,"score":0.2770000100135803},{"id":"https://openalex.org/C18294631","wikidata":"https://www.wikidata.org/wiki/Q2621185","display_name":"Location parameter","level":3,"score":0.27489998936653137},{"id":"https://openalex.org/C175291020","wikidata":"https://www.wikidata.org/wiki/Q1156822","display_name":"Offset (computer science)","level":2,"score":0.2727000117301941},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.27239999175071716},{"id":"https://openalex.org/C110121322","wikidata":"https://www.wikidata.org/wiki/Q865811","display_name":"Distribution (mathematics)","level":2,"score":0.2630000114440918},{"id":"https://openalex.org/C16345878","wikidata":"https://www.wikidata.org/wiki/Q107472979","display_name":"Orientation (vector space)","level":2,"score":0.26179999113082886},{"id":"https://openalex.org/C144386022","wikidata":"https://www.wikidata.org/wiki/Q1332997","display_name":"Scale factor (cosmology)","level":5,"score":0.25920000672340393},{"id":"https://openalex.org/C113346285","wikidata":"https://www.wikidata.org/wiki/Q6804193","display_name":"Measured depth","level":2,"score":0.25369998812675476}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.21203/rs.3.rs-10799029/v1","doi":"https://doi.org/10.21203/rs.3.rs-10799029/v1","is_oa":true,"landing_page_url":"https://doi.org/10.21203/rs.3.rs-10799029/v1","pdf_url":"https://www.researchsquare.com/article/rs-10799029/latest.pdf","source":{"id":"https://openalex.org/S4306525896","display_name":"Research Square","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"acceptedVersion","is_accepted":true,"is_published":false,"raw_source_name":"Research Square","raw_type":"posted-content"}],"best_oa_location":{"id":"doi:10.21203/rs.3.rs-10799029/v1","doi":"https://doi.org/10.21203/rs.3.rs-10799029/v1","is_oa":true,"landing_page_url":"https://doi.org/10.21203/rs.3.rs-10799029/v1","pdf_url":"https://www.researchsquare.com/article/rs-10799029/latest.pdf","source":{"id":"https://openalex.org/S4306525896","display_name":"Research Square","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"acceptedVersion","is_accepted":true,"is_published":false,"raw_source_name":"Research Square","raw_type":"posted-content"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7225129488.pdf","grobid_xml":"https://content.openalex.org/works/W7225129488.grobid-xml"},"referenced_works_count":26,"referenced_works":["https://openalex.org/W1903029394","https://openalex.org/W1923184257","https://openalex.org/W2609822318","https://openalex.org/W2752782242","https://openalex.org/W2775906317","https://openalex.org/W2776622059","https://openalex.org/W2789541106","https://openalex.org/W3015871565","https://openalex.org/W3100969133","https://openalex.org/W3194841206","https://openalex.org/W3207804100","https://openalex.org/W3217020749","https://openalex.org/W4206789359","https://openalex.org/W4241881238","https://openalex.org/W4255822925","https://openalex.org/W4288080166","https://openalex.org/W4312372834","https://openalex.org/W4312725821","https://openalex.org/W4384648484","https://openalex.org/W4386065698","https://openalex.org/W4386179772","https://openalex.org/W4402818207","https://openalex.org/W4402916775","https://openalex.org/W4403791547","https://openalex.org/W4413155942","https://openalex.org/W7125801993"],"related_works":["https://openalex.org/W4413155942","https://openalex.org/W7154329056","https://openalex.org/W7117337176","https://openalex.org/W2963896186","https://openalex.org/W2803442140","https://openalex.org/W4407606071","https://openalex.org/W4403828432","https://openalex.org/W7208742866","https://openalex.org/W7011735930","https://openalex.org/W2960933010"],"abstract_inverted_index":{"Abstract":[0],"Purpose:":[1],"RGB-D":[2],"semantic":[3],"segmentation":[4],"uses":[5,76,223],"depth":[6,26,32,40,57,61,80,92,121,125,141,226,230],"information":[7],"to":[8,28,82,150,163,235],"reduce":[9],"appearance":[10],"ambiguity.":[11],"However,":[12],"geometric":[13,49,161],"prior":[14],"weights":[15],"shared":[16],"across":[17],"inputs":[18],"cannot":[19],"adjust":[20],"the":[21,66,89,131,139,153,159,164,207,224],"contributions":[22],"of":[23,133,194,232],"spatial":[24,90],"and":[25,59,91,124,157,168,174,200,215,228],"cues":[27,231],"individual":[29],"scenes.":[30],"Pairwise":[31],"differences":[33],"also":[34],"do":[35],"not":[36],"explicitly":[37],"represent":[38],"local":[39,60,120,229],"variation.":[41],"This":[42],"work":[43],"introduces":[44],"a":[45,107],"mechanism":[46],"that":[47],"calibrates":[48],"biases":[50,93],"for":[51,88,110],"each":[52,77,95,99,111,128,233],"input":[53,234],"using":[54],"its":[55,134],"relative":[56,79,225],"distribution":[58,81,227],"cues.":[62],"Methods:":[63],"We":[64],"propose":[65],"Reliability-aware":[67],"Geometry-Prior":[68],"Transformer":[69],"(RGPFormer).":[70],"Adaptive":[71],"Geometry":[72],"Prior":[73],"Gating":[74],"(AGPG)":[75],"sample's":[78],"generate":[83],"separate":[84],"overall":[85],"scaling":[86,108,149],"factors":[87,137],"in":[94,113,152],"attention":[96,115,237],"head":[97,116],"at":[98],"stage.":[100],"Local":[101],"Depth":[102,198],"Reliability":[103],"Modulation":[104],"(LDRM)":[105],"derives":[106],"factor":[109],"position":[112,129],"every":[114],"from":[117],"normalized":[118],"depth,":[119],"standard":[122],"deviation,":[123],"gradient.":[126],"For":[127],"pair,":[130],"product":[132],"two":[135],"endpoint":[136],"modulates":[138],"corresponding":[140],"bias.":[142],"The":[143,179],"full":[144],"configuration":[145,181],"applies":[146],"asymmetric":[147],"residual":[148],"AGPG":[151,173],"final":[154],"encoder":[155],"stage":[156],"adds":[158],"calibrated":[160],"bias":[162],"scores":[165,193,212],"between":[166],"queries":[167],"keys":[169],"before":[170],"Softmax.":[171],"Results:":[172],"LDRM":[175],"add":[176],"0.033M":[177],"parameters.":[178],"evaluated":[180],"based":[182],"on":[183,196,202],"DFormerv2-S,":[184],"denoted":[185],"RGPFormer-S,":[186],"achieved":[187],"mean":[188],"intersection":[189],"over":[190],"union":[191],"(mIoU)":[192],"58.25%":[195],"NYU":[197],"V2":[199],"52.16%":[201],"SUN":[203],"RGB-D.":[204],"Compared":[205],"with":[206,238],"matched":[208],"baseline":[209],"A0,":[210],"these":[211],"were":[213],"0.94":[214],"0.49":[216],"percentage":[217],"points":[218],"higher,":[219],"respectively.":[220],"Conclusion:":[221],"RGPFormer":[222],"adapt":[236],"low":[239],"parameter":[240],"overhead.":[241]},"counts_by_year":[],"updated_date":"2026-11-04T10:54:05.112534","created_date":"2026-11-03T00:00:00"}