AIS. AI-in-Sensor. No digitizing. No latency . No distortion. No bus. No power -- almost. No problem.
AIS. AI-in-Sensor. No digitizing. No latency . No distortion. No bus. No power -- almost. No problem.
A storm is coming. AIStorm presents its AI-in-Sensor SoCs …
… Sensors are not digital devices. To convert sensor information into a digital representation takes time, power, silicon, and adds distortion & noise. AIS solutions can be finished processing before the competition even has enough digital information to begin. So why not skip the middle man and consider our SoCs for your next project?
Technology
We’ve all heard about process in memory. Why stop there?
The sensors are the systems eyes, ears & fingers. Sensors can focus on what we care about and ignore the rest – a face, another car, an intruder, a sound, a change…
…event driven digital just isn’t & memory needs something to remember. Consider an image sensor watching a warehouse. Only a few pixels are involved with the area around a door in the far corner where an intruder might enter.
So the system has to poll, or incorporate some complicated handshake scheme with the processing system to try to work out where to focus. No matter how it is done it is slow and complicated. So why bother? Now consider a camera scanning faces at an intersection. Distractions abound… cars, trees, clouds, winds, rain, snow, debris blowing by. How is a digital system going to deal with that with all those ADCs and handshakes going on?
AIS technology is analog and mixed signal technology but its more than that. For the first time the sensor itself is smart enough to deal with events, can make decisions, and can perform complex analysis… often before the other guy even receives his digital sensor representation to start processing.
Imagine a smart car braking faster; or a camera lasting years on a single battery; biometric login by small wearable devices like watches, bands and glasses. Imagine enabling new technologies like handset gesture control, heart based security, through screen image sensors and advanced machine vision.
The Nitty Gritty
- compatible with TensorFlow & other AI abstraction languages
- programmable architecture
- >10 TOPs/W
- >2.5 TOPs
- up to 100’s of thousands of neurons in a few square mm’s
- Mhz to Multi-GHz DEC* clocking speeds
- distributed wavefront memory
- fast deep learning capable
- no noise penalty
- integrated sensors
- low cost
*Digital Equivalent
Applications
Through Screen Fingerprint Sensor
Augmented Transportation
Security & Traffic Cameras
A Warehouse Security Camera
A digital system watching the door in the background could easily miss an intruder due to the need to continuously poll different areas of the image. Otherwise, complicated interface schemes are required. AIS offers true event driven operation that can allow years of operation on a single battery.
Digital systems need complex arbiters to ignore background information like trees, cars, rain, snow, wind, clouds, debris, etc. and focus on faces. With AIS the cost and complication of digital processing is not needed.
A digital system watching the door in the background could easily miss an intruder due to the need to continuously poll different areas of the image. Otherwise, complicated interface schemes are required. AIS offers true event driven operation that can allow years of operation on a single battery.
A Face Recognition Traffic Camera
Security & Traffic Cameras
A Warehouse Security Camera
A digital system watching the door in the background could easily miss an intruder due to the need to continuously poll different areas of the image. Otherwise, complicated interface schemes are required. AIS offers true event driven operation that can allow years of operation on a single battery.
A Face Recognition Traffic Camera
Digital systems need complex arbiters to ignore background information like trees, cars, rain, snow, wind, clouds, debris, etc. and focus on faces. With AIS the cost and complication of digital processing is not needed.
A digital system watching the door in the background could easily miss an intruder due to the need to continuously poll different areas of the image. Otherwise, complicated interface schemes are required. AIS offers true event driven operation that can allow years of operation on a single battery.
"Heart Muscle Waveform Authentication" vs. "Fingerprint"
Wearable devices require a continuous, long life, authentication scheme. AIS and its partner companies own the patents to this technology and are using AIS to make it work.
Gesture control
Gesture control sensors can operate in the 50mW range, but the processing presently takes too much power for handsets or watches. AIS technology enables this new technology making it usable in portable applications.
Food service automation
Freeing personnel from redundant tasks and lowering costs for consumers, AI is already redefining the food industry. AIS and its partners are leaders in the food equipment industry and will help usher in the next generation of AI powered products.
AI medical imaging diagnosis
AI augmented medical machine diagnostics is helping the medical industry reduce errors, distribute leading techniques, and bring their services to remote locations. AIS technology can help improve medical devices, improving performance and lowering costs.
Edge voice processing
Voice is a key human interface device (HID). Already cell phones and stand-alone devices offer high-performance voice solutions. However, they are often dependent upon the cloud for operation. By pushing more capability into the device, bandwidth requirements can be reduced and the technology can be used for transportation and other applications where the cloud connection may not be reliable.
In the News:
https://www.biometricupdate.
https://www.iotevolutionworld.
https://pulse2.com/aistorm-
https://venturebeat.com/2019/
Careers
AIStorm is hiring in San Jose, Phoenix, Graz, Dresden, Hsinchu & Austin:
• Analog IC Designers
• Digital IC Designers
• IC Verification Engineers
• IC Layout
• Test Engineering
• Algorithm Engineers
• Business Managers
Please send your resume to:
Responsibilities
We’re growing our team -- if you’re a high-energy, self-starting engineer with hands on experience with Artificial Intelligence and good understanding of Analog applications, AIStorm is the place for you.
AIS technology is analog and mixed signal technology, where the sensor does all the decision making and perform complex analysis rather than performing a digital sensor processing and then start the decision making.
-Develop solutions for real-world, using Artificial Intelligence/Machine Learning; and help team minimize the parameters and design the best network for audio and image applications.
-This role bridges the gap between Analog Engineer and a Software Developer.
-Develop key techniques for Audio and Image applications in domains such as audio recognition, Computer vision, etc. and interdisciplinary domains..
Qualifications
-MS or PhD degree in Computer Science, Artificial Intelligence, Machine Learning, or related technical field with understanding in Analog Circuits.
-Ideal candidate will be skilled in audio and image processing, and machine learning applications-Working experience with Deep Learning using Tensor Flow, Cafe.
-Hands on Python, Matlab experience.
-Technically oriented, proactive, and enthusiastic, with extreme attention to detail.
About Us
Our backers include a group of more than forty investors, including one of the world’s leading suppliers of authentication systems, one of the world’s largest CIS image sensor manufacturers, a leading food service equipment supplier owning many of the world’s best known brands, a leading wearable OEM company and many others.
Our investors are also our customers.
Our investors are leaders in their fields offering AIStorm unique access to properly define solutions and the technology needed to outperform competitors.
Our team is comprised of engineering, manufacturing and business leaders from the semiconductor industry. Our team has collectively produced more than one thousand SoCs for industry leading semiconductor companies. We know how to define, develop and manufacture microchips.
AIStorm is based in San Jose, CA with offices in Austria, Taiwan, Phoenix and soon Dresden and Israel.
Why invest in AIStorm?
