Context encoder guided self-supervised siamese depth estimation based on stereo laparoscopic images.
Wenda LiYuichiro HayashiMasahiro OdaTakayuki KitasakaKazunari MisawaKensaku MoriPublished in: Medical Imaging: Image-Guided Procedures (2021)
Keyphrases
- depth estimation
- stereo vision
- depth map
- binocular stereo
- stereo matching
- multi view stereo
- dense depth maps
- depth cues
- depth perception
- disparity map
- depth information
- stereo algorithms
- stereo method
- model based pose estimation
- depth estimates
- three dimensional
- stereo pair
- scene understanding
- image data
- input image
- real scenes
- object recognition
- ground truth
- dynamic scenes
- feature matching
- multi view
- stereo reconstruction
- motion parallax
- depth from defocus
- stereo images
- image registration
- super resolution
- bit rate
- edge detection
- input data
- feature points
- space time
- image matching
- specular reflection
- d scene
- position and orientation