FID value comparison between the real dataset of 1000 real images and the synthetic datasets of 1000 synthetic images generated from different GAN architectures which are modified to generate four channels outputs.
dataset
posted on 2022-05-02, 17:40 authored by Vajira Thambawita, Pegah Salehi, Sajad Amouei Sheshkal, Steven A. Hicks, Hugo L. Hammer, Sravanthi Parasa, Thomas de Lange, Pål Halvorsen, Michael A. RieglerFID value comparison between the real dataset of 1000 real images and the synthetic datasets of 1000 synthetic images generated from different GAN architectures which are modified to generate four channels outputs.
History
Usage metrics
Categories
Keywords
requiring tremendous effortscorresponding ground truthprepare synthetic imagesevaluated using qualitativecorresponding masks usingpipeline significantly improvesdiv >< panalyzing medical datasingle training imagesynthetic data generatedseg pipeline improvessingle imagesynthetic dataperformed usinggenerated datatraining datamedical datamedical expertsseg pipelinereal datadata samplestraining datasetssupportive toolsegmentation modelsseg rare abnormalitiesquantitative comparisonspublicly availableprivacy reasonspopular toolopen datasetmodel needsmedicine duelarge amountsinfrequent lesionshighly dependentgeneral lackfind abnormalitiescostly taskconsiderable amountclose performancesautomatic processingartificial intelligence
Licence
Exports
RefWorks
BibTeX
Ref. manager
Endnote
DataCite
NLM
DC