Large-scale aerial scene perception based on self-supervised multi-view stereo via cycled generative adversarial network.
Kevin W. TongZhiyi ShiGuangyu ZhuYa DuanYuhong HouEdmond Qi WuLimin ZhuPublished in: Inf. Fusion (2024)
Keyphrases
- multi view stereo
- multi view
- depth map
- urban environments
- depth estimation
- structure from motion
- scene reconstruction
- stereo reconstruction
- bundle adjustment
- ray tracing
- d scene
- point cloud
- three dimensional
- camera parameters
- high fidelity
- scene flow
- depth information
- camera motion
- single image
- input image
- video sequences