Yesterday, Japanese electronics manufacturer Sony announced its formal entry to the field of AI-powered drones with the launch of Airpeak. Today, Silicon Valley-based startup Mipsology announced that it has partnered with the Japanese design and development company OKI IDS to bring hardware solutions for AI. Specifically, the two firms will be collaborating on the AI-powered image processing market in Japan.
OKI IDS will be leveraging the power of Mipsology's Zebra ML inference accelerator, which is a deep learning compute engine for neural network inference. Zebra is one of the first technologies that speed up inference by replacing or complementing CPU/GPUs with field-programmable gate arrays (FPGAs) to accelerate ML workflows and underlying computations. FGPAs do all this at lower power requirements, better environmental adaptability, with a longer lifespan than most GPUs.
Following today's announcement, OKI IDS will be combining Zebra's current rapport of 14 FGPAs boards with its own expertise in designing complex FGPA-based systems to bring effective ML solutions to the market. “Mipsology is the perfect partner to help us expand our ML development support services on FPGAs,” said Satoshi Shimizu, President and CEO, OKI IDS. “We expect the Japanese market for high-speed systems integrating ML and other functions into FPGAs to grow rapidly and substantially. FPGAs have no competition when it comes to adaptability, high performance, and high quality.”
“We are excited by our collaboration with OKI IDS to supply the Japanese demand for neural network computation that supports high-speed, low-latency image processing,” commented Ludovic Larzul, CEO and founder of Mipsology. “OKI IDS and Mipsology have a natural synergy and mutual expertise in FPGA design and machine learning that will result in our delivering industry-leading products based on FPGAs.”
The two companies hope that this collaboration will cater to the increasing usage of AI across a wide variety of markets in Japan, including autonomous driving, industrial robotics, smart cities, medical applications, and video monitoring.