2021 BRN Discussion, page-12280

Hi MC,

From your FRAMOS link, it doesn't quack like Akida and it doesn't waddle Akida.

SLVS-EC’s Advantages Over MIPI CSI-2 / D-PHY
Spearhead for Sony’s future image sensor technology
Embedded clock increases robustness and signal integrity
Simplified designs, lane-to-lane skew is less critical
Longer routing tracks are possible due to more robust signalling
Up to 8 lanes and double the bandwidth of a single sensor to 20 Gbps

Until now, this technology was limited to FPGA professionals and industrial applications due to its complex nature and engineering knowledge required to implement it correctly. Now, FRAMOS has created an opportunity for NVIDIA Jetson developers to benefit from these high-performance Sony sensors with their SLVS-EC interfaces. Customers can fully exploit the 2.5 Gbps data throughput on all 8 lanes to get a native overall bandwidth of almost 20 Gbps from a single image sensor within their specific imaging solution. This opens them up to a much broader sensor selection, specifically ones developed for the industrial or AV market.

They are talking about vary large bit-rates - Akida is in the business of economising on data rates.

https://www.framos.com/en/news/vision-insights-finding-the-right-vision-system-for-every-drone
How Artificial Intelligence Serves Object Recognition
Deep learning algorithms or artificial intelligence provide more security for drones through more precise collision avoidance and/or enable autonomous tracking of objects and persons. For collision avoidance, the drone must detect obstacles such as walls, trees or other objects independently and in real time; and perform precise evasive maneuvers. During tracking tasks, the drone automatically detects the object of interest and can follow it automatically.

For these tasks, neural networks train the software to recognize objects. At least 10,000 images, sometimes up to several million, are required to provide reliable test data for machine learning on a high-performance computer such as a GPU. The results of these computations allow object recognition algorithms to run on a small, energy-efficient processor architecture, such as an ASIC chip.