GANVO: Unsupervised Deep Monocular Visual Odometry and Depth Estimation with Generative Adversarial Networks.
Yasin AlmaliogluMuhamad Risqi U. SaputraPedro Porto Buarque de GusmãoAndrew MarkhamNiki TrigoniPublished in: ICRA (2019)
Keyphrases
- visual odometry
- depth estimation
- ego motion
- depth map
- autonomous navigation
- depth images
- depth information
- stereo vision
- dynamic scenes
- stereo matching
- stereo camera
- optical flow
- real scenes
- simultaneous localization and mapping
- feature matching
- image sequences
- scene understanding
- stereo pair
- real time
- d scene
- camera pose
- high quality
- semi supervised
- viewpoint
- object recognition
- space time