針對圖像指代分割的訓(xùn)練后量化策略
打開文本圖片集
關(guān)鍵詞：圖像指代分割；訓(xùn)練后量化；跨模態(tài)融合；深度學(xué)習(xí)
中圖分類號：TP391.41;TP183 文獻(xiàn)標(biāo)志碼：A 文章編號：1001-3695（2025）07-014-2025-07
doi：10.19734/j. issn.1001-3695.2024.10.0437
Abstract：RISaims tosegmentobjectsdescribedbysentencesinanimagebyintegratingvisualandlinguisticinformation. This technique has strong appication prospects ininteractiveimage editingandlanguage-guided human-machine interaction. However，existing solutions tendtoexplore high-performance models，neglecting practicalapplicationsonedgedeviceswith limited esources.ThepaperproposedaneficientPQframework toaddressthischallenge.Specifically，theanalysisdentifiedtherotcauseofperformancecollpsecausedbyusingtheround-to-nearest（RTN）quantization method.Thentheframework proposedatwo-regionbalancedquantizationstrategytosolvethenon-normaldistributionofactivationvaluesaftersoftmax and GELUoperations inthevisual encoder，andintroducedareordered groupingquantization strategytotacklethequantizationproblemscausedbyoutliersinthelinearlayersactivationof the textencoder.Extensiveexperimentsexploringdierent quantization bitwidthsonthreebenchmark datasetsdemonstratetheclearadvantages ofthe proposed methodover existing PTQ methods.AsthefirstquantizationschemespecificallfortheRIStask，itconfirmsthefeasibilityofeficientlydeployingthe RIS model to edge devices using the PTQ method.
Key words： referring image segmentation（RIS）； post-training quantization（PTQ）;cross-model fusion; deep learning
0引言
深度學(xué)習(xí)極大程度提高了視覺算法在許多圖像分割任務(wù)上的性能，如語義分割[1]實(shí)例分割[2]等。（剩余18265字）