1. SceneNet RGB-D: can 5M synthetic images beat generic ImageNet pre-training on indoor segmentation?;McCormac,2017
2. ScanNet: Richly-annotated 3D reconstructions of indoor scenes;Dai,2017
3. Fully convolutional networks for semantic segmentation;Long,2015
4. Semantic image segmentation with deep convolutional nets and fully connected CRFs;Chen,2015
5. Multi-scale context aggregation by dilated convolutions;Yu,2016