QTTNet: Quantized tensor train neural networks for 3D object and video recognition.
Donghyun LeeDingheng WangYukuan YangLei DengGuangshe ZhaoGuoqi LiPublished in: Neural Networks (2021)
Keyphrases
- d objects
- object recognition
- neural network
- occluded objects
- invariant recognition
- pattern recognition
- three dimensional objects
- recognition and pose estimation
- object representation
- viewpoint
- recognizing objects
- multi view
- pose estimation
- multiple views
- range data
- video sequences
- three dimensional
- cad models
- video frames
- shape descriptors
- d mesh
- video data
- object matching
- object classes
- space time
- spin images
- shape recognition
- physically plausible
- object features
- machine learning
- image contours
- object model
- line drawings
- object detection
- computer vision
- static images
- object models
- object class
- multi views