Object segmentation in depth maps with one user click and a synthetically trained fully convolutional network.
Matthieu GrardRomain BrégierFlorian SellaEmmanuel DellandréaLiming ChenPublished in: CoRR (2018)
Keyphrases
- object segmentation
- convolutional network
- depth map
- convolutional neural networks
- coarse to fine
- object detection
- multi view
- image segmentation
- graph cuts
- high quality
- segmentation algorithm
- three dimensional
- segmentation method
- image sequences
- shape prior
- input image
- object boundaries
- object categories
- high resolution
- object classes
- multiscale
- low resolution
- super resolution
- computer vision
- image features
- object recognition
- human computer interaction
- information retrieval
- image registration
- keywords