Matches in SemOpenAlex for { <https://semopenalex.org/work/W2899202481> ?p ?o ?g. }
Showing items 1 to 96 of
96
with 100 items per page.
- W2899202481 abstract "We present a method that meta classifies whether seg-ments predicted by a semantic segmentation neural networkintersect with the ground truth. For this purpose, we employ measures of dispersion for predicted pixel-wise class probability distributions, like classification entropy, that yield heat maps of the input scene's size. We aggregate these dispersion measures segment-wise and derive metrics that are well-correlated with the segment-wise IoU of prediction and ground truth. This procedure yields an almost plug and play post-processing tool to rate the prediction quality of semantic segmentation networks on segment level. This is especially relevant for monitoring neural networks in online applications like automated driving or medical imaging where reliability is of utmost importance. In our tests, we use publicly available state-of-the-art networks trained on the Cityscapes dataset and the BraTS2017 dataset and analyze the predictive power of different metrics as well as different sets of metrics. To this end, we compute logistic LASSO regression fits for the task of classifying IoU=0 vs. IoU>0 per segment and obtain AUROC values of up to 91.55%. We complement these tests with linear regression fits to predict the segment-wise IoU and obtain prediction standard deviations of down to 0.130 as well as $R^2$ values of up to 84.15%. We show that these results clearly outperform standard approaches." @default.
- W2899202481 created "2018-11-09" @default.
- W2899202481 creator A5014806681 @default.
- W2899202481 creator A5029365974 @default.
- W2899202481 creator A5036830968 @default.
- W2899202481 creator A5054573957 @default.
- W2899202481 creator A5065666645 @default.
- W2899202481 creator A5080542566 @default.
- W2899202481 creator A5083693267 @default.
- W2899202481 date "2018-11-01" @default.
- W2899202481 modified "2023-09-28" @default.
- W2899202481 title "Prediction Error Meta Classification in Semantic Segmentation: Detection via Aggregated Dispersion Measures of Softmax Probabilities" @default.
- W2899202481 cites W1641498739 @default.
- W2899202481 cites W1901129140 @default.
- W2899202481 cites W1976526581 @default.
- W2899202481 cites W1993947467 @default.
- W2899202481 cites W1995875735 @default.
- W2899202481 cites W2103328396 @default.
- W2899202481 cites W2123402141 @default.
- W2899202481 cites W2135046866 @default.
- W2899202481 cites W2340897893 @default.
- W2899202481 cites W2480078828 @default.
- W2899202481 cites W2531327146 @default.
- W2899202481 cites W2531409750 @default.
- W2899202481 cites W2624172162 @default.
- W2899202481 cites W2751069891 @default.
- W2899202481 cites W2783000019 @default.
- W2899202481 cites W2803756472 @default.
- W2899202481 cites W2884490794 @default.
- W2899202481 cites W2914617358 @default.
- W2899202481 cites W2963046541 @default.
- W2899202481 cites W2964059111 @default.
- W2899202481 cites W2964309882 @default.
- W2899202481 cites W2989692526 @default.
- W2899202481 hasPublicationYear "2018" @default.
- W2899202481 type Work @default.
- W2899202481 sameAs 2899202481 @default.
- W2899202481 citedByCount "1" @default.
- W2899202481 countsByYear W28992024812020 @default.
- W2899202481 crossrefType "posted-content" @default.
- W2899202481 hasAuthorship W2899202481A5014806681 @default.
- W2899202481 hasAuthorship W2899202481A5029365974 @default.
- W2899202481 hasAuthorship W2899202481A5036830968 @default.
- W2899202481 hasAuthorship W2899202481A5054573957 @default.
- W2899202481 hasAuthorship W2899202481A5065666645 @default.
- W2899202481 hasAuthorship W2899202481A5080542566 @default.
- W2899202481 hasAuthorship W2899202481A5083693267 @default.
- W2899202481 hasConcept C105795698 @default.
- W2899202481 hasConcept C124101348 @default.
- W2899202481 hasConcept C146849305 @default.
- W2899202481 hasConcept C153083717 @default.
- W2899202481 hasConcept C153180895 @default.
- W2899202481 hasConcept C154945302 @default.
- W2899202481 hasConcept C188441871 @default.
- W2899202481 hasConcept C33923547 @default.
- W2899202481 hasConcept C41008148 @default.
- W2899202481 hasConcept C50644808 @default.
- W2899202481 hasConcept C83546350 @default.
- W2899202481 hasConcept C89600930 @default.
- W2899202481 hasConceptScore W2899202481C105795698 @default.
- W2899202481 hasConceptScore W2899202481C124101348 @default.
- W2899202481 hasConceptScore W2899202481C146849305 @default.
- W2899202481 hasConceptScore W2899202481C153083717 @default.
- W2899202481 hasConceptScore W2899202481C153180895 @default.
- W2899202481 hasConceptScore W2899202481C154945302 @default.
- W2899202481 hasConceptScore W2899202481C188441871 @default.
- W2899202481 hasConceptScore W2899202481C33923547 @default.
- W2899202481 hasConceptScore W2899202481C41008148 @default.
- W2899202481 hasConceptScore W2899202481C50644808 @default.
- W2899202481 hasConceptScore W2899202481C83546350 @default.
- W2899202481 hasConceptScore W2899202481C89600930 @default.
- W2899202481 hasOpenAccess W2899202481 @default.
- W2899202481 hasRelatedWork W1614999412 @default.
- W2899202481 hasRelatedWork W1760702393 @default.
- W2899202481 hasRelatedWork W2050921384 @default.
- W2899202481 hasRelatedWork W2064460620 @default.
- W2899202481 hasRelatedWork W2157011304 @default.
- W2899202481 hasRelatedWork W2742248641 @default.
- W2899202481 hasRelatedWork W2766915612 @default.
- W2899202481 hasRelatedWork W2803756472 @default.
- W2899202481 hasRelatedWork W2902210009 @default.
- W2899202481 hasRelatedWork W2945027741 @default.
- W2899202481 hasRelatedWork W2953196973 @default.
- W2899202481 hasRelatedWork W2970291460 @default.
- W2899202481 hasRelatedWork W2970523552 @default.
- W2899202481 hasRelatedWork W3041775211 @default.
- W2899202481 hasRelatedWork W3090238777 @default.
- W2899202481 hasRelatedWork W3100707783 @default.
- W2899202481 hasRelatedWork W3108363523 @default.
- W2899202481 hasRelatedWork W3128194091 @default.
- W2899202481 hasRelatedWork W3194082517 @default.
- W2899202481 hasRelatedWork W2560147515 @default.
- W2899202481 isParatext "false" @default.
- W2899202481 isRetracted "false" @default.
- W2899202481 magId "2899202481" @default.
- W2899202481 workType "article" @default.