INV-Flow2PoseNet: Light-Resistant Rigid Object Pose from Optical Flow of RGB-D Images using Images, Normals and Vertices.
Torben FetzerGerd ReisDidier StrickerPublished in: CoRR (2022)
Keyphrases
- optical flow
- three dimensional
- ground truth
- rigid body
- flow field
- image features
- input image
- test images
- image retrieval
- pose estimation
- bounding box
- object recognition
- computer vision
- image classification
- color images
- color information
- object pose
- segmentation algorithm
- segmentation method
- vector field
- region of interest
- motion segmentation
- illumination conditions
- high resolution