Frustum ConvNet: Sliding Frustums to Aggregate Local Point-Wise Features for Amodal 3D Object Detection.
Zhixin WangKui JiaPublished in: CoRR (2019)
Keyphrases
- d objects
- pointwise
- surface features
- object features
- multi view
- object recognition
- viewpoint
- image features
- shape descriptors
- three dimensional
- range data
- d mesh
- object model
- partial matching
- object representation
- weighted sum
- occluded objects
- multiple views
- surface properties
- three dimensional objects
- pose estimation
- feature space
- multi views
- topological information
- physically plausible
- object contours
- salient features
- object classes
- object detection
- small number