Matches in SemOpenAlex for { <https://semopenalex.org/work/W4312583896> ?p ?o ?g. }
Showing items 1 to 94 of
94
with 100 items per page.
- W4312583896 endingPage "9" @default.
- W4312583896 startingPage "262933" @default.
- W4312583896 abstract "Semantic image segmentation plays a crucial role in a wide range of industrial applications and has been receiving significant attention. Unfortunately, image segmentation tasks are notoriously difficult and different industries often require human experts. Convolutional neural networks (CNNs) have been successfully applied in many fields of image segmentation. But all of them still require a huge amount of hand-labeled data for training. A lot of research was conducted in the field of unsupervised and semi-supervised learning, which studies how to shrink the amount of training data at the same time preserving the quality of the model. But still another field of research - transformation of “cheap” (in terms of time, money and human resources) markup into “expensive” is novel. In this work a new approach of generating semantic segmentation masks, using only classification labels of the image, was proposed. Proposed method is based on the GradCam algorithm, which can produce image activation heatmap, using only class label. But GradCams’ heatmaps are raw for final use, so additional techniques and transforms should be applied in order to get final usable masks. Experiments were conducted on the task of detecting defects on steel plates — Kaggle- Severstal: Steel Defect Detection. After that Dice metric was computed using a classical training approach and proposed method: classical approach - 0.621, proposed method - 0.465. Proposed approach requires much less human resources compared to the classical approach. Moreover, after visual inspection of results it is obvious that the proposed approach has successfully completed the task of defect localization." @default.
- W4312583896 created "2023-01-05" @default.
- W4312583896 creator A5016068553 @default.
- W4312583896 date "2022-08-25" @default.
- W4312583896 modified "2023-10-10" @default.
- W4312583896 title "Method of Transformation of Image Classification Labels into Segmentation Masks" @default.
- W4312583896 cites W2765793020 @default.
- W4312583896 cites W2949533892 @default.
- W4312583896 cites W2952339589 @default.
- W4312583896 cites W2962834855 @default.
- W4312583896 cites W2964121744 @default.
- W4312583896 cites W3091002423 @default.
- W4312583896 cites W3102564565 @default.
- W4312583896 cites W3125832420 @default.
- W4312583896 cites W3162276117 @default.
- W4312583896 cites W3167724130 @default.
- W4312583896 cites W3216067289 @default.
- W4312583896 cites W4214592090 @default.
- W4312583896 cites W4221154584 @default.
- W4312583896 cites W4248083651 @default.
- W4312583896 cites W4287025617 @default.
- W4312583896 cites W4288089157 @default.
- W4312583896 cites W4288364646 @default.
- W4312583896 doi "https://doi.org/10.20535/2523-4455.mea.262933" @default.
- W4312583896 hasPublicationYear "2022" @default.
- W4312583896 type Work @default.
- W4312583896 citedByCount "0" @default.
- W4312583896 crossrefType "journal-article" @default.
- W4312583896 hasAuthorship W4312583896A5016068553 @default.
- W4312583896 hasBestOaLocation W43125838961 @default.
- W4312583896 hasConcept C104317684 @default.
- W4312583896 hasConcept C108583219 @default.
- W4312583896 hasConcept C115961682 @default.
- W4312583896 hasConcept C119857082 @default.
- W4312583896 hasConcept C124504099 @default.
- W4312583896 hasConcept C136764020 @default.
- W4312583896 hasConcept C153180895 @default.
- W4312583896 hasConcept C154945302 @default.
- W4312583896 hasConcept C162324750 @default.
- W4312583896 hasConcept C185592680 @default.
- W4312583896 hasConcept C187736073 @default.
- W4312583896 hasConcept C202444582 @default.
- W4312583896 hasConcept C204241405 @default.
- W4312583896 hasConcept C2780451532 @default.
- W4312583896 hasConcept C2780615836 @default.
- W4312583896 hasConcept C33923547 @default.
- W4312583896 hasConcept C41008148 @default.
- W4312583896 hasConcept C55493867 @default.
- W4312583896 hasConcept C81363708 @default.
- W4312583896 hasConcept C89600930 @default.
- W4312583896 hasConcept C9652623 @default.
- W4312583896 hasConceptScore W4312583896C104317684 @default.
- W4312583896 hasConceptScore W4312583896C108583219 @default.
- W4312583896 hasConceptScore W4312583896C115961682 @default.
- W4312583896 hasConceptScore W4312583896C119857082 @default.
- W4312583896 hasConceptScore W4312583896C124504099 @default.
- W4312583896 hasConceptScore W4312583896C136764020 @default.
- W4312583896 hasConceptScore W4312583896C153180895 @default.
- W4312583896 hasConceptScore W4312583896C154945302 @default.
- W4312583896 hasConceptScore W4312583896C162324750 @default.
- W4312583896 hasConceptScore W4312583896C185592680 @default.
- W4312583896 hasConceptScore W4312583896C187736073 @default.
- W4312583896 hasConceptScore W4312583896C202444582 @default.
- W4312583896 hasConceptScore W4312583896C204241405 @default.
- W4312583896 hasConceptScore W4312583896C2780451532 @default.
- W4312583896 hasConceptScore W4312583896C2780615836 @default.
- W4312583896 hasConceptScore W4312583896C33923547 @default.
- W4312583896 hasConceptScore W4312583896C41008148 @default.
- W4312583896 hasConceptScore W4312583896C55493867 @default.
- W4312583896 hasConceptScore W4312583896C81363708 @default.
- W4312583896 hasConceptScore W4312583896C89600930 @default.
- W4312583896 hasConceptScore W4312583896C9652623 @default.
- W4312583896 hasIssue "2" @default.
- W4312583896 hasLocation W43125838961 @default.
- W4312583896 hasLocation W43125838962 @default.
- W4312583896 hasLocation W43125838963 @default.
- W4312583896 hasOpenAccess W4312583896 @default.
- W4312583896 hasPrimaryLocation W43125838961 @default.
- W4312583896 hasRelatedWork W1522196789 @default.
- W4312583896 hasRelatedWork W2982321410 @default.
- W4312583896 hasRelatedWork W3029198973 @default.
- W4312583896 hasRelatedWork W3133861977 @default.
- W4312583896 hasRelatedWork W3167935049 @default.
- W4312583896 hasRelatedWork W3193565141 @default.
- W4312583896 hasRelatedWork W4226493464 @default.
- W4312583896 hasRelatedWork W4312417841 @default.
- W4312583896 hasRelatedWork W4315434538 @default.
- W4312583896 hasRelatedWork W95465806 @default.
- W4312583896 hasVolume "27" @default.
- W4312583896 isParatext "false" @default.
- W4312583896 isRetracted "false" @default.
- W4312583896 workType "article" @default.