Plugging Self-Supervised Monocular Depth into Unsupervised Domain Adaptation for Semantic Segmentation.
Adriano CardaceLuca De LuigiPierluigi Zama RamirezSamuele SaltiLuigi di StefanoPublished in: CoRR (2021)
Keyphrases
- semantic segmentation
- street scenes
- superpixels
- conditional random fields
- label transfer
- weakly supervised
- scene classification
- image sequences
- depth information
- object classes
- depth estimation
- depth map
- object class
- pascal voc
- object categories
- object segmentation
- pixel level
- image understanding
- natural images
- object detection
- probabilistic model
- high resolution
- object recognition