Depth from Monocular Images using a Semi-Parallel Deep Neural Network (SPDNN) Hybrid Architecture.
Shabab BazrafkanHossein JavidniaJoe LemleyPeter CorcoranPublished in: CoRR (2017)
Keyphrases
- hybrid architecture
- monocular images
- depth estimation
- stereo vision
- depth map
- human body
- pose estimation
- stereo matching
- depth information
- process management
- dynamic scenes
- optical flow
- agent architecture
- stereo pair
- video sequences
- human pose
- image sequences
- three dimensional
- software development
- object detection
- information systems
- scene understanding
- real scenes
- depth images