SynSense in China has boosted its recent $10m funding round to support its 3D neuromorphic processor technology developed in Switzerland.
The latest round raising ‘tens of millions of RMB’ was led by Chinese investors Maxvision and RunWoo. The funds will be used to accelerate the development of SynSense’s 3D neuromorphic processor DYNAP-CNN2 for high-speed vision processing for autonomous systems, such as drones, robots and self-driving vehicles.
- SynSense raises US$10 million for smart vision sensor
- Swiss neuromorphic startup changes name, raises money
- Prophesee, SynSense form vision computing partnership
The DYNAP-CNN2 processor handles depth estimation, corner detection, optical flow and ego motion computing modules, based on dynamic vision processing. The new chip will support complex visual applications such as ultra-low power autonomous flight and obstacle avoidance.
This follows a $10m round announced last week, led by Ausvic Capital, to accelerate the mass production of its current Speck chip. The Speck chip combines SynSense’s low-power spiking neural network (SNN) vision processor with an event-based sensor to perform both neuromorphic sensing and computing using a fully asynchronous circuit design. It can be used to capture real-time visual information, recognize and detect objects, and perform other vision-based detection and interaction functions.
The neuromorphic technology was originally developed at the Institute of Neuroinformatics of the University of Zurich and the ETH Zurich and spun out as aiCTX in March 2017 in Zürich. It has been headquartered in Chengdu since 2020, retaining advanced R&D in Zürich and adding research in Shenzen and chip design in Shanghai.
“As a high-tech enterprise dedicated to AI and industry integration, Maxvision has long been focused on cutting-edge AI-related technologies. SynSense’s neuromorphic technology is highly valuable in edge computing scenarios because of its ultra-low power consumption and ultra-low latency characteristics,” said Lei Qu, Chairman of Maxvision Technology, a publicly listed enterprise on the Main Board of the Shenzhen Stock Exchange, headquartered in Shenzhen, Guangdong. “By combining SynSense’s technology with our overall solutions design, we aim to deliver industry-leading solutions for various intelligent scenarios.”
According to Pengyu Lu, founding partner of venture capital fund RunWoo in Guangdong, Hong Kong, and Macao, AI is moving in two main directions: complex solutions in the cloud, requiring high power consumption; and ultra-low power consumption, low-latency applications on the edge.
“Low-power sensors and processors are the key to the popularization of AIoT. Spiking neural networks and asynchronous circuits are natural ultra-low power architectures. SynSense has demonstrated global technology leadership and talent reserves in this field, successfully bringing their neuromorphic chips to mass production. We believe SynSense will be a major player in the edge intelligence sector.”
Other Synsense articles
- Spiking neural network dev kits for sound and vision
- Swiss-Chinese AI chip developer cuts cockpit deal with BMW
- Europe spends €3 million in neuromorphic R&D
SynSense is working with German car maker BMW on the Speck chip. “Speck doesn’t need caches or additional cameras. It can capture visual event information, process real-time information computing, and conduct smart scene analysis with less than 1 milliwatt of power consumption and with 5-10 millisecond end-to-end latency,” said Ning Qiao, founder and CEO of SynSense.