Red Light, Green Light and Tailgate Detection Among Latest Features Demoed at CES 2024
Syntiant Corp., a leader in edge AI deployment, announced a series of new algorithms to its lineup of highly accurate, high performance machine learning models to enable smarter and more efficient vehicles.
Analyzing multiple acoustic and image characteristics in real time through advanced AI and machine learning, Syntiant has developed and trained specialized edge AI models that enhance vehicle safety and security, while adding several new features including:
- Red light and green light detection. Determines if a vehicle is stopped at a red light or accelerating at a green light.
- Tailgating detection. Determines if vehicles are following too closely, particularly at higher speeds. Leverages accelerometer and GPS inputs to determine vehicle motion and speed.
- Glass-break detection. Determines if a car window is broken.
- Person, vehicle and license plate detection. Detects key objects while the vehicle is parked, keeping items of interest at full resolution, compressing the rest of the image, saving storage space on the device.
Recommended AI News: SiLC Technologies Unveils the Eyeonic Vision System Mini: A Leap in Precision LiDAR Technology
“We are introducing our latest advancements for the automotive sector as demand grows among manufacturers to deploy smart technology features that boost overall vehicle safety and security,” said Kurt Busch, CEO of Syntiant. “Whether it is noise suppression, voice commands, blind spot detection or facial recognition, our highly accurate, production-ready models, along with our ultra-low-power Neural Decision Processors, provide a complete turnkey solution for OEMs and developers to bring advanced features to vehicles that improve battery life, privacy and user experiences, all at significantly lower cost.”
Recommended AI News: Alibaba Boosts SMEs With New AI Feature at CES 2024
Moving Deep Learning from the Cloud to the Edge
With Syntiant’s technology, many machine learning applications, which previously could only be implemented on cloud servers or high-powered processors, can now run in a low-power, always-on domain at the edge. Syntiant’s proprietary model architectures enable world-leading inference speed and minimized memory footprint across a broad range of hardware platforms, including CPUs, GPUs, DSPs, FPGAs and ASICs. The company’s Neural Decision Processors have been independently verified to be 100x more power efficient and offer 10x the throughput when compared to existing low-power MCUs. The combination is powering larger networks, while consuming significantly less power.
Whether it is an acoustic event detector for security applications, advanced video processing in a teleconferencing device or dash cam, or real time monitoring of battery health, Syntiant provides developers and integrators with high performance, proven solutions that can take them from concept to product in the shortest time possible.
Recommended AI News: SignalSTAT: AI-Boosted Veterinary Radiology for Faster, Better Patient Care
[To share your insights with us, please write to sghosh@martechseries.com]
Comments are closed.