6 papers
Multi-domain semantic segmentation with overlapping labels
Petra Bevandić, Marin Oršić, Ivan Grubišić +2
Deep supervised models have an unprecedented capacity to absorb large quantities of training data. Hence, training on many datasets becomes a method of choice towards graceful degr…
Dense open-set recognition with synthetic outliers generated by Real NVP
Matej Grcić, Petra Bevandić, Siniša Šegvić
Today's deep models are often unable to detect inputs which do not belong to the training distribution. This gives rise to confident incorrect predictions which could lead to devas…
Simultaneous Semantic Segmentation and Outlier Detection in Presence of Domain Shift
Petra Bevandić, Ivan Krešo, Marin Oršić +1
Recent success on realistic road driving datasets has increased interest in exploring robust performance in real-world applications. One of the major unsolved problems is to identi…
In Defense of Pre-trained ImageNet Architectures for Real-time Semantic Segmentation of Road-driving Images
Marin Oršić, Ivan Krešo, Petra Bevandić +1
Recent success of semantic segmentation approaches on demanding road driving datasets has spurred interest in many related application fields. Many of these applications involve re…
Discriminative out-of-distribution detection for semantic segmentation
Petra Bevandić, Ivan Krešo, Marin Oršić +1
Most classification and segmentation datasets assume a closed-world scenario in which predictions are expressed as distribution over a predetermined set of visual classes. However,…
Robust Semantic Segmentation with Ladder-DenseNet Models
Ivan Krešo, Marin Oršić, Petra Bevandić +1
We present semantic segmentation experiments with a model capable to perform predictions on four benchmark datasets: Cityscapes, ScanNet, WildDash and KITTI. We employ a ladder-sty…