Array Camera Image Fusion using Physics-Aware Transformers
This addresses image synthesis challenges for camera arrays with varied specifications, but appears incremental as it builds on existing transformer methods.
The paper tackled the problem of fusing images from diverse camera arrays by introducing a physics-aware transformer, achieving scalable synthetic training data generation using open-source software.
We demonstrate a physics-aware transformer for feature-based data fusion from cameras with diverse resolution, color spaces, focal planes, focal lengths, and exposure. We also demonstrate a scalable solution for synthetic training data generation for the transformer using open-source computer graphics software. We demonstrate image synthesis on arrays with diverse spectral responses, instantaneous field of view and frame rate.