An Adversarial Generative Network Designed for High-Resolution Monocular Depth Estimation from 2D HiRISE Images of Mars.
Riccardo La GrassaIgnazio GalloCristina ReGabriele CremoneseNicola LandroClaudio PernecheleEmanuele SimioniMattia GattiPublished in: Remote. Sens. (2022)
Keyphrases
- depth estimation
- monocular images
- depth cues
- depth map
- depth perception
- high resolution
- input image
- stereo vision
- three dimensional
- model based pose estimation
- ground truth
- image features
- image data
- image analysis
- super resolution
- depth information
- image retrieval
- depth estimates
- stereo matching
- object recognition
- low resolution
- image processing
- multi view
- dynamic scenes
- high quality
- scene understanding
- pose estimation
- input data
- real scenes
- geometric constraints