English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

We present a method that "meta" classifies whether seg-ments predicted by asemantic segmentation neural networkintersect with the ground truth. For thispurpose, we employ measures of dispersion for predicted pixel-wise classprobability distributions, like classification entropy, that yield heat maps ofthe input scene's size. We aggregate these dispersion measures segment-wise andderive metrics that are well-correlated with the segment-wise IoU of predictionand ground truth. This procedure yields an almost plug and play post-processingtool to rate the prediction quality of semantic segmentation networks onsegment level. This is especially relevant for monitoring neural networks inonline applications like automated driving or medical imaging where reliabilityis of utmost importance. In our tests, we use publicly availablestate-of-the-art networks trained on the Cityscapes dataset and the BraTS2017dataset and analyze the predictive power of different metrics as well asdifferent sets of metrics. To this end, we compute logistic LASSO regressionfits for the task of classifying IoU=0 vs. IoU>0 per segment and obtain AUROCvalues of up to 91.55%. We complement these tests with linear regression fitsto predict the segment-wise IoU and obtain prediction standard deviations ofdown to 0.130 as well as $R^2$ values of up to 84.15%. We show that theseresults clearly outperform standard approaches.

Prediction Error Meta Classification in Semantic Segmentation: Detection via Aggregated Dispersion Measures of Softmax Probabilities