SegICP: Integrated Deep Semantic Segmentation and Pose Estimation.
Jay Ming WongVincent KeeTiffany LeSyler WagnerGian Luca MariottiniAbraham SchneiderLei HamiltonRahul ChipalkattyMitchell HebertDavid M. S. JohnsonJimmy WuBolei ZhouAntonio TorralbaPublished in: CoRR (2017)
Keyphrases
- pose estimation
- semantic segmentation
- d objects
- conditional random fields
- superpixels
- computer vision
- object classes
- human body
- scene classification
- object categories
- feature points
- object class
- weakly supervised
- object recognition
- viewpoint
- image set
- training examples
- multi view
- higher order
- image features
- hidden markov models
- feature vectors
- video sequences
- multiscale