GTA: A Geometry-Aware Attention Mechanism for Multi-View Transformers.
Takeru MiyatoBernhard JaegerMax WellingAndreas GeigerPublished in: CoRR (2023)
Keyphrases
- multi view
- attention mechanism
- three dimensional
- multi view stereo
- view dependent
- visual attention
- multiple views
- single view
- depth map
- d objects
- camera calibration
- saliency map
- semi supervised
- visual attention model
- multiple cameras
- multi view learning
- multi view face detection
- surface reconstruction
- range images
- view synthesis
- geometric constraints
- real time
- multiple viewpoints
- eye tracking
- multi view clustering
- human motion
- free viewpoint
- multi view images
- high level