Efficiera is an ultra-low power AI inference accelerator that can be implemented on an FPGA device or ASIC/ASSP device, and is specialized for CNN inference operations.
Efficiera addresses various technical challenges including power consumption, cost, and heat dissipation, enabling the rapid introduction of on-device edge AI products to the market.
Efficiera deep learning models are pre-trained models that can be deployed on edge devices. We have developed and provide pre-trained models for object detection, anomaly detection, and noise reduction.
Click here for more information about our extremely low bit quantization technology.
Utilization with FPGAs
Utilization with ASIC/ASSP
Efficiera achieves both excellent power and area efficiency, allowing energy savings and cost reduction for edge devices running AI solutions.
データの移動量とビット数を最小化することで、コンボリューションの演算処理に要する電力を削減します。
演算ロジック削減により、演算サイクル数を削減でき、面積・周波数あたりの演算性能を向上させます。
演算ビット数を最小化することにより、演算器1個あたりの回路面積とSRAMサイズを最小化します。
回路構成の選択により演算性能を調整できるため、お客様の実施したいタスクに合わせて、コンフィギュレーションを最適化し、Efficieraの性能をあげることが可能です。
黎明期からソフトウェアとハードウェアの両面から研究開発を続けてきたモデルの軽量化技術や
専用回路設計技術・知見などをパッケージングして製品化します。
The advantage of introducing Efficiera with FPGAs is the reduction of BoM cost owing to the small size of the IP itself.
By implementing Efficiera in the free space of an existing FPGA design or on a larger device with compatible pins, AI capabilities can be added without changing the number of components. It can also be mounted on small FPGAs, such as Cyclone® V SoCs.
Efficiera can be suitable for various situations across every industry.
Improve image quality by removing noise caused by low light conditions and compression artifacts caused by ima ...
Obscure personal information such as faces when data is collected. ...
Detect cracks in concrete. ...
Detect specific dangerous behaviors, such as wandering and falls, from skeletal information. ...
Count large numbers of people automatically. ...
Issue a warning when people approach a surrounding obstacle. ...
Efficiera can be implemented on a wide varaiety of. hardware platforms from different manufacturers.
We are excited about the shortened development cycles and reduced development costs this will enable for the mass production of on-device AI solutions.
KEIm-CVSoC, developed by Kondo Electronics Industry Co., Ltd., is a SoM (system on module) equipped with Intel® Cyclon® V FPGAs. They are suited to a wide range of applications from technological studies to product prototyping to mass production.
The So-One Module, developed by PALTEK, is the world's smallest module equipped with an MPSoC. The MPSoC, from Xilinx, is well-suited to edge computing.
We work together with strong partner companies to achieve practical applications and mass production of edge AI devices for our customers.
With the participation of experts in various fields with extensive knowledge of FPGAs, we fully utilize technologies and knowledge based on our abundant experiences to enable practical applications and mass production of edge AI devices for our customers.
