Our Binary Pixel solution was inspired from photographic need. Rambus imaging team-members are all deeply passionate about photography and videography. When not working at engineering technology solutions, we can often be found photographing the world around us. We have often noticed that many photos and videos simply do not match real-life for high contrast scenes typical in daily life, such as bright landscapes, sunset portraits, and scenes with both sunlight and shadow. The problem stems from the image sensor in current cameras which captures only a fraction of the dynamic range that the human eye can see. In varying light conditions, the camera must make a compromise by choosing which portion of the scene, bright or dark details, to capture. The results are usually dark or washed-out images with crushed blacks and blown out highlights.
The imaging team set out to solve this problem by applying our knowledge of chip design, signal processing & managed dataflow learned from our 100+ years of combined experience designing sensors and camera systems. Our goal is to create a technology that captures the full gamut of dark to bright intensities in the real-world while simplifying the exposure algorithm of the camera to take the guess work out of photography. After all, the photographic experience is about capturing “keeper” photos and videos that showcase images as seen by our human eyes.
While architecting binary pixel technology, we are fortunate to collaborate with key experts in industry and academia. Binary Pixel technology builds upon the visionary works of imaging and signal processing experts including: The Gigavision Camera by Professor Martin Vetterli at École Polytechnique Fédérale de Lausanne (EPFL) and Professor Edoardo Charbon at Delft University of Technology & EPFL; The Digital Film Sensor by Dr. Eric Fossum, pioneer in the modern CMOS active pixel image sensor.
The physics behind our solution centers on the pixel’s full well capacity, its ability to collect photo-electrons into an electronic “bucket” during the exposure period. The smaller the pixel (such as in smartphone cameras), the smaller the “bucket” and the lower the dynamic range. Binary pixels employ advanced oversampling at the point-of-capture on the sensor, resulting in an order of magnitude higher dynamic range than a conventional CMOS image sensor pixel. The implementation can be manufactured using current CMOS technology, and fits in a comparable form-factor, cost and power envelope as current image sensors.
We believe binary pixel technology can play a key part in enabling the world’s best sensor and camera systems. We look forward to working with the industry to integrate this technology into future mobile and consumer cameras.
More about the current Binary Pixel Architecture can be found in the following presentation.
Links:
Binary Pixel at rambus.com
“Overcoming the Full Well Capacity Limit: High Dynamic Range Imaging Using Multi-Bit Temporal Oversampling and Conditional Reset”
Paper | Presentation
Leave a Reply