CEA Leti's new autonomous imager uses 'much less energy' than conventional cameras

France-based research agency CEA-Leti has announced what it describes as "the world's first autonomous imager technology that activates smartphones and small appliances through face recognition or other specific patterns."

Called μWAI (micro-WAY) and the same size as a €1 coin, the autonomous imager features a novel readout and processing architecture co-designed with an optimized algorithmic pipeline.

The recognition and recording of data starts with a sequence of elementary algorithms, making for ultralow-power wake-up modes. The design is based on a compact silicon format to keep costs down.

CEA Leti says the μWAI "is the first smart image sensor jointly featuring auto-exposure for all lighting conditions and 88dB dynamic range, as well as motion detection and feature extraction for event-based functioning, and AI-based object recognition that triggers highly reliable identification."

'Energy demand 10,000 times less'

These key features also enable reliable decision-making for a few tens of pJ / pixel / frame, which outperforms conventional systems, say the developers. The pJ/pixel/frame measures the energy spent by each pixel for each single image within a frame of images. A typical [conventional] implementation requires about 10,000 times more energy than μWAI, says CEA Leti.

The autonomous imager is designed for everyday small appliances. Its key features include:

• Energy efficiency: consuming 10k times less than low-power camera-plus-processor set;
• Privacy-compliant, AI-based recognition: nearly human-detection performance (95 percent) ;
• Wide operating lighting sensitivity to ensure accurate recognition in highly variable conditions;
• Five-year, always-on CR1025 battery lifetime;
• 3-6μW operation, typically required for Internet of Things apps; and
• Works with button cell that lasts five years

Applications

Applications and functions of the new sensor include automatic switching and face identification in mobile devices, contactless control of "smart" household appliances and sport-and-entertainment devices.

The μWAI also enables face recognition, people counting, alarm triggering in smart buildings, vehicle-interior situation awareness, driver identification, parking-situation awareness and a smart-unlocking system in automobiles.

Antoine Dupret, CEA-Leti's industrial partnership manager, commented, "The μWAI's recognition engine is optimized to recognize faces when movement is detected. CEA-Leti's team is working in partnership with STMicroelectronics to develop specific smart-imager products and we are considering extending the technology to other use cases. We target adapting the recognition engine as IP embedded in various cameras and optimizing the performance of the imager to the requirements of our partner's customers."

The technology behind μWAI will be introduced during CEA-Leti's flagship digital event Leti Innovation Days, June 22 and 23, 2021.