GlocalFuse-Depth: Fusing transformers and CNNs for all-day self-supervised monocular depth estimation.
Zezheng ZhangRyan K. Y. ChanKenneth K. Y. WongPublished in: Neurocomputing (2024)
Keyphrases
- depth estimation
- monocular images
- depth cues
- depth map
- stereo vision
- depth perception
- depth information
- depth estimates
- stereo matching
- dynamic scenes
- dense depth maps
- stereo pair
- scene understanding
- stereo camera
- image sequences
- super resolution
- real scenes
- pose estimation
- high resolution
- multi view
- disparity map
- d scene
- vision system
- feature matching
- object detection
- depth from defocus
- epipolar geometry
- image processing
- computer vision
- real time