Researchers at KAUST (King Abdullah University of Science and Technology) have presented Fovea Stacking, a computational imaging technique built around Phaseform's Deformable Phase Plate (DPP). Shi Mao, Dr. Yogeshwar Nath Mishra and Prof. Wolfgang Heidrich presented the work at ACM SIGGRAPH Asia 2025 in Hong Kong, 15 to 18 December 2025.
The idea
A simple lens produces sharp images only near the centre of the field; towards the edges, aberrations blur the picture. Fovea Stacking borrows a strategy from the human eye, which only resolves fine detail in a small region, the fovea, and moves that region around. A DPP in front of the lens corrects the aberrations locally for one region of interest at a time. A stack of such captures is then merged into a single image that is sharp across the entire field of view.
The result is dynamic, localized aberration correction in real time, which simplifies the optical system, reduces size and complexity, and improves image quality at the same time.
Where it could go
The authors point to autonomous vehicles and navigation, VR and AR displays with foveated rendering, surveillance and security imaging, satellite and aerial imaging, and extended depth-of-field imaging as potential applications.
The full paper, videos and code are on the Fovea Stacking project page. We will show Fovea Stacking live at VISION 2026 in Stuttgart. If you are developing next-generation computational cameras, we offer evaluation kits; get in touch via our contact page.
