eGAC3D: enhancing depth adaptive convolution and depth estimation for monocular 3D object pose detection.
Duc-Tuan NgoMinh-Quan Viet BuiDuc Dung NguyenHoang-Anh PhamPublished in: PeerJ Comput. Sci. (2022)
Keyphrases
- d objects
- depth estimation
- monocular images
- depth map
- pose estimation
- multi view
- depth cues
- stereo vision
- depth information
- object recognition
- three dimensional
- scene understanding
- dynamic scenes
- stereo matching
- depth estimates
- multiple views
- viewpoint
- object detection
- range data
- d scene
- stereo pair
- stereo camera
- real scenes
- depth images
- image sequences
- human body
- human pose
- surface reconstruction
- super resolution
- position and orientation
- high quality
- disparity map
- depth from defocus
- image processing
- feature matching
- object boundaries
- vision system
- camera calibration
- high resolution
- stereo images
- computer vision