Exploiting Motion Perception in Depth Estimation Through a Lightweight Convolutional Neural Network.
Pedro Nuno LeiteAndry Maykol PintoPublished in: IEEE Access (2021)
Keyphrases
- lightweight
- depth estimation
- convolutional neural network
- depth cues
- dynamic scenes
- stereo vision
- depth map
- depth estimates
- stereo matching
- image sequences
- face detection
- space time
- depth information
- motion estimation
- motion model
- scene understanding
- moving objects
- high quality
- super resolution
- stereo pair
- real scenes
- optical flow
- camera motion
- computer vision
- motion field
- wireless sensor networks
- neural network
- multi view
- feature matching
- color information
- video sequences
- specular reflection
- d scene
- image matching