EPNet: Enhancing Point Features with Image Semantics for 3D Object Detection.
Tengteng HuangZhe LiuXiwu ChenXiang BaiPublished in: ECCV (15) (2020)
Keyphrases
- d objects
- point features
- multiple views
- image features
- object recognition
- input image
- multi view
- shape descriptors
- viewpoint
- pose estimation
- point correspondences
- reference images
- single image
- camera pose
- three dimensional
- single view
- hough transform
- structure from motion
- high resolution
- range data
- position and orientation
- range images
- ground plane
- multiple images
- object classes
- line drawings
- computer vision
- medical images
- image segmentation