Scene captioning with deep fusion of images and point clouds.
Qiang YuChunxia ZhangLubin WengShiming XiangChunhong PanPublished in: Pattern Recognit. Lett. (2022)
Keyphrases
- point cloud
- multi view stereo
- laser scanner
- real world objects
- input image
- multiple images
- three dimensional
- architectural scenes
- urban scenes
- structure from motion
- image registration
- image data
- surface reconstruction
- rigid registration
- point features
- image regions
- geometric constraints
- complex scenes
- stereo camera
- image sequences
- geometric information
- uncalibrated images
- video sequences
- uncalibrated cameras
- stereo reconstruction
- icp algorithm
- urban environments
- viewpoint
- cad model
- image processing
- similarity measure
- normal vectors
- depth estimation
- point sets
- single image