Matches in SemOpenAlex for { <https://semopenalex.org/work/W2970291460> ?p ?o ?g. }
Showing items 1 to 80 of
80
with 100 items per page.
- W2970291460 abstract "In the semantic segmentation of street scenes the reliability of the prediction and therefore uncertainty measures are of highest interest. We present a method that generates for each input image a hierarchy of nested crops around the image center and presents these, all re-scaled to the same size, to a neural network for semantic segmentation. The resulting softmax outputs are then post processed such that we can investigate mean and variance over all image crops as well as mean and variance of uncertainty heat maps obtained from pixel-wise uncertainty measures, like the entropy, applied to each crop's softmax output. In our tests, we use the publicly available DeepLabv3+ MobilenetV2 network (trained on the Cityscapes dataset) and demonstrate that the incorporation of crops improves the quality of the prediction and that we obtain more reliable uncertainty measures. These are then aggregated over predicted segments for either classifying between IoU = 0 and IoU > 0 (meta classification) or predicting the IoU via linear regression (meta regression). The latter yields reliable performance estimates for segmentation networks, in particular useful in the absence of ground truth. For the task of meta classification we obtain a classification accuracy of 81.93% and an AUROC of 89.89%. For meta regression we obtain an R <sup xmlns:mml=http://www.w3.org/1998/Math/MathML xmlns:xlink=http://www.w3.org/1999/xlink>2</sup> value of 84.77%. These results yield significant improvements compared to other approaches." @default.
- W2970291460 created "2019-09-05" @default.
- W2970291460 creator A5028586330 @default.
- W2970291460 creator A5080542566 @default.
- W2970291460 date "2019-06-01" @default.
- W2970291460 modified "2023-10-01" @default.
- W2970291460 title "Uncertainty Measures and Prediction Quality Rating for the Semantic Segmentation of Nested Multi Resolution Street Scene Images" @default.
- W2970291460 cites W1976526581 @default.
- W2970291460 cites W2123402141 @default.
- W2970291460 cites W2137493099 @default.
- W2970291460 cites W2284860901 @default.
- W2970291460 cites W2340897893 @default.
- W2970291460 cites W2480078828 @default.
- W2970291460 cites W2803756472 @default.
- W2970291460 cites W2884490794 @default.
- W2970291460 cites W2964309882 @default.
- W2970291460 doi "https://doi.org/10.1109/cvprw.2019.00176" @default.
- W2970291460 hasPublicationYear "2019" @default.
- W2970291460 type Work @default.
- W2970291460 sameAs 2970291460 @default.
- W2970291460 citedByCount "23" @default.
- W2970291460 countsByYear W29702914602019 @default.
- W2970291460 countsByYear W29702914602020 @default.
- W2970291460 countsByYear W29702914602021 @default.
- W2970291460 countsByYear W29702914602022 @default.
- W2970291460 countsByYear W29702914602023 @default.
- W2970291460 crossrefType "proceedings-article" @default.
- W2970291460 hasAuthorship W2970291460A5028586330 @default.
- W2970291460 hasAuthorship W2970291460A5080542566 @default.
- W2970291460 hasBestOaLocation W29702914602 @default.
- W2970291460 hasConcept C105795698 @default.
- W2970291460 hasConcept C119857082 @default.
- W2970291460 hasConcept C121955636 @default.
- W2970291460 hasConcept C124101348 @default.
- W2970291460 hasConcept C144133560 @default.
- W2970291460 hasConcept C146849305 @default.
- W2970291460 hasConcept C153180895 @default.
- W2970291460 hasConcept C154945302 @default.
- W2970291460 hasConcept C188441871 @default.
- W2970291460 hasConcept C196083921 @default.
- W2970291460 hasConcept C33923547 @default.
- W2970291460 hasConcept C41008148 @default.
- W2970291460 hasConcept C48921125 @default.
- W2970291460 hasConcept C50644808 @default.
- W2970291460 hasConcept C83546350 @default.
- W2970291460 hasConcept C89600930 @default.
- W2970291460 hasConceptScore W2970291460C105795698 @default.
- W2970291460 hasConceptScore W2970291460C119857082 @default.
- W2970291460 hasConceptScore W2970291460C121955636 @default.
- W2970291460 hasConceptScore W2970291460C124101348 @default.
- W2970291460 hasConceptScore W2970291460C144133560 @default.
- W2970291460 hasConceptScore W2970291460C146849305 @default.
- W2970291460 hasConceptScore W2970291460C153180895 @default.
- W2970291460 hasConceptScore W2970291460C154945302 @default.
- W2970291460 hasConceptScore W2970291460C188441871 @default.
- W2970291460 hasConceptScore W2970291460C196083921 @default.
- W2970291460 hasConceptScore W2970291460C33923547 @default.
- W2970291460 hasConceptScore W2970291460C41008148 @default.
- W2970291460 hasConceptScore W2970291460C48921125 @default.
- W2970291460 hasConceptScore W2970291460C50644808 @default.
- W2970291460 hasConceptScore W2970291460C83546350 @default.
- W2970291460 hasConceptScore W2970291460C89600930 @default.
- W2970291460 hasLocation W29702914601 @default.
- W2970291460 hasLocation W29702914602 @default.
- W2970291460 hasOpenAccess W2970291460 @default.
- W2970291460 hasPrimaryLocation W29702914601 @default.
- W2970291460 hasRelatedWork W158826679 @default.
- W2970291460 hasRelatedWork W2151356126 @default.
- W2970291460 hasRelatedWork W2743258233 @default.
- W2970291460 hasRelatedWork W2921182884 @default.
- W2970291460 hasRelatedWork W2938746851 @default.
- W2970291460 hasRelatedWork W2997969508 @default.
- W2970291460 hasRelatedWork W3208883981 @default.
- W2970291460 hasRelatedWork W4211165872 @default.
- W2970291460 hasRelatedWork W4307834408 @default.
- W2970291460 hasRelatedWork W4320925816 @default.
- W2970291460 isParatext "false" @default.
- W2970291460 isRetracted "false" @default.
- W2970291460 magId "2970291460" @default.
- W2970291460 workType "article" @default.