{"id":"https://openalex.org/W4413278147","doi":"https://doi.org/10.1109/icip55913.2025.11084527","title":"SLICE: Synthetic Caption-Trained Lightweight Image Captioner for Edge Devices","display_name":"SLICE: Synthetic Caption-Trained Lightweight Image Captioner for Edge Devices","publication_year":2025,"publication_date":"2025-08-18","ids":{"openalex":"https://openalex.org/W4413278147","doi":"https://doi.org/10.1109/icip55913.2025.11084527"},"language":"en","primary_location":{"id":"doi:10.1109/icip55913.2025.11084527","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icip55913.2025.11084527","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE International Conference on Image Processing (ICIP)","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5074549508","display_name":"Shayan Joya","orcid":null},"institutions":[{"id":"https://openalex.org/I4210117523","display_name":"Samsung (United Kingdom)","ror":"https://ror.org/01w6gjq94","country_code":"GB","type":"company","lineage":["https://openalex.org/I2250650973","https://openalex.org/I4210117523"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Shayan Joya","raw_affiliation_strings":["Samsung Research,UK"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Samsung Research,UK","institution_ids":["https://openalex.org/I4210117523"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5052069148","display_name":"Soroush Fatemifar","orcid":"https://orcid.org/0000-0003-3460-1093"},"institutions":[{"id":"https://openalex.org/I4210117523","display_name":"Samsung (United Kingdom)","ror":"https://ror.org/01w6gjq94","country_code":"GB","type":"company","lineage":["https://openalex.org/I2250650973","https://openalex.org/I4210117523"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Soroush Fatemifar","raw_affiliation_strings":["Samsung Research,UK"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Samsung Research,UK","institution_ids":["https://openalex.org/I4210117523"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5055524671","display_name":"Ali Akbari","orcid":"https://orcid.org/0000-0003-1457-2977"},"institutions":[{"id":"https://openalex.org/I4210117523","display_name":"Samsung (United Kingdom)","ror":"https://ror.org/01w6gjq94","country_code":"GB","type":"company","lineage":["https://openalex.org/I2250650973","https://openalex.org/I4210117523"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Ali Akbari","raw_affiliation_strings":["Samsung Research,UK"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Samsung Research,UK","institution_ids":["https://openalex.org/I4210117523"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Andrew Sheppard","orcid":null},"institutions":[{"id":"https://openalex.org/I4210117523","display_name":"Samsung (United Kingdom)","ror":"https://ror.org/01w6gjq94","country_code":"GB","type":"company","lineage":["https://openalex.org/I2250650973","https://openalex.org/I4210117523"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Andrew Sheppard","raw_affiliation_strings":["Samsung Research,UK"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Samsung Research,UK","institution_ids":["https://openalex.org/I4210117523"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5110781208","display_name":"Christopher Alder","orcid":null},"institutions":[{"id":"https://openalex.org/I4210117523","display_name":"Samsung (United Kingdom)","ror":"https://ror.org/01w6gjq94","country_code":"GB","type":"company","lineage":["https://openalex.org/I2250650973","https://openalex.org/I4210117523"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Christopher Alder","raw_affiliation_strings":["Samsung Research,UK"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Samsung Research,UK","institution_ids":["https://openalex.org/I4210117523"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I4210117523"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.20974113,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1864","last_page":"1869"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10627","display_name":"Advanced Image and Video Retrieval Techniques","score":0.9983999729156494,"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/T10627","display_name":"Advanced Image and Video Retrieval Techniques","score":0.9983999729156494,"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":0.9886000156402588,"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/T11605","display_name":"Visual Attention and Saliency Detection","score":0.9883999824523926,"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/computer-science","display_name":"Computer science","score":0.6929547190666199},{"id":"https://openalex.org/keywords/enhanced-data-rates-for-gsm-evolution","display_name":"Enhanced Data Rates for GSM Evolution","score":0.6297976970672607},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5024750232696533},{"id":"https://openalex.org/keywords/computer-graphics","display_name":"Computer graphics (images)","score":0.4955838918685913},{"id":"https://openalex.org/keywords/computer-vision","display_name":"Computer vision","score":0.45626023411750793},{"id":"https://openalex.org/keywords/image","display_name":"Image (mathematics)","score":0.45149725675582886}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6929547190666199},{"id":"https://openalex.org/C162307627","wikidata":"https://www.wikidata.org/wiki/Q204833","display_name":"Enhanced Data Rates for GSM Evolution","level":2,"score":0.6297976970672607},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5024750232696533},{"id":"https://openalex.org/C121684516","wikidata":"https://www.wikidata.org/wiki/Q7600677","display_name":"Computer graphics (images)","level":1,"score":0.4955838918685913},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.45626023411750793},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.45149725675582886}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icip55913.2025.11084527","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icip55913.2025.11084527","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE International Conference on Image Processing (ICIP)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":18,"referenced_works":["https://openalex.org/W1861492603","https://openalex.org/W1956340063","https://openalex.org/W2481240925","https://openalex.org/W2506483933","https://openalex.org/W2904565150","https://openalex.org/W2963016445","https://openalex.org/W2964616647","https://openalex.org/W3091588028","https://openalex.org/W3139732141","https://openalex.org/W3193290865","https://openalex.org/W4312289196","https://openalex.org/W4312671791","https://openalex.org/W4312922092","https://openalex.org/W4372337842","https://openalex.org/W4382460199","https://openalex.org/W4385574358","https://openalex.org/W4386076004","https://openalex.org/W4386076674"],"related_works":["https://openalex.org/W2772917594","https://openalex.org/W2036807459","https://openalex.org/W2058170566","https://openalex.org/W2755342338","https://openalex.org/W2166024367","https://openalex.org/W3116076068","https://openalex.org/W2229312674","https://openalex.org/W2951359407","https://openalex.org/W2079911747","https://openalex.org/W1969923398"],"abstract_inverted_index":{"Image":[0,56],"captioning":[1,13,30,165],"has":[2],"made":[3],"significant":[4],"steps":[5],"forward":[6],"in":[7,124,132],"recent":[8],"years.":[9],"Existing":[10],"state-of-the-art":[11,37,64],"image":[12,29,164],"models":[14,38,166],"are":[15],"heavily":[16],"parameterized":[17],"and":[18,43,104,136],"a":[19,44,68,82,109,150,183],"long":[20],"way":[21],"from":[22],"being":[23],"deployable":[24],"on":[25,32,171],"edge":[26,33],"devices.":[27],"Providing":[28],"inference":[31],"devices":[34,60],"with":[35,167],"current":[36],"requires":[39],"expensive":[40],"server-side":[41],"infrastructure":[42],"stable":[45],"network":[46],"connection.":[47],"In":[48],"this":[49,79],"paper,":[50],"we":[51],"propose":[52],"Synthetic":[53],"caption-trained":[54],"Lightweight":[55],"Captioning":[57],"for":[58],"Edge":[59],"(SLICE).":[61],"It":[62],"provides":[63],"caption":[65,117],"quality":[66,143],"at":[67],"slice":[69],"of":[70,73,127,186],"the":[71,86,116,125,146],"compute":[72],"existing":[74,163],"large":[75],"models.":[76],"We":[77],"achieve":[78],"by":[80,89,107],"proposing":[81],"lightweight":[83],"pipeline,":[84],"redesigning":[85],"baseline":[87,147],"architecture":[88],"making":[90],"multiple":[91],"modifications.":[92],"The":[93],"model":[94,128,142,148],"is":[95],"first":[96],"pre-trained":[97],"using":[98,149],"our":[99,179],"own":[100],"generated":[101],"synthetic":[102],"captions":[103],"then":[105],"fine-tuned":[106],"adopting":[108],"novel":[110],"knowledge":[111],"distillation":[112],"approach":[113],"to":[114,145],"improve":[115],"quality.":[118],"Our":[119,157],"small-SLICE":[120,181],"achieves":[121,159,182],"8.9x":[122],"reduction":[123,131],"number":[126],"parameters,":[129],"7.9x":[130],"peak":[133],"memory":[134],"consumption":[135],"10.5x":[137],"latency":[138,185],"improvement,":[139],"while":[140],"surpassing":[141],"compared":[144],"smaller":[151],"input":[152],"size":[153],"achieving":[154],"138.9":[155],"CIDEr.":[156],"large-SLICE":[158],"147.3":[160],"CIDEr":[161],"outperforming":[162],"only":[168],"35.8M":[169],"parameters":[170],"COCO":[172],"Karpathy":[173],"test":[174],"split.":[175],"On":[176],"desktop":[177],"CPU,":[178],"proposed":[180],"low":[184],"36ms.":[187]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2025-10-10T00:00:00"}