Weakly Supervised Training of Universal Visual Concepts for Multi-domain Semantic Segmentation

Dense Out-of-Distribution Detection by Robust Learning on Synthetic Negative Data

DenseHybrid: Hybrid Anomaly Detection for Dense Open-Set Recognition

Dense open-set recognition based on training with noisy negative images

Automatic universal taxonomies for multi-domain semantic segmentation

Multi-domain semantic segmentation with overlapping labels

Dense open-set recognition with synthetic outliers generated by Real NVP

Multi-domain semantic segmentation with pyramidal fusion

Traffic Scene Classification on a Representation Budget

Simultaneous Semantic Segmentation and Outlier Detection in Presence of Domain Shift

In Defense of Pre-Trained ImageNet Architectures for Real-Time Semantic Segmentation of Road-Driving Images

Convolutional Models for Segmentation and Localization