Vast amounts of AI development are currently poured into cloud computation. These solutions are based on a mix of CPU, GPU, VPU, TPU, FPGA, ASIC and SoC technologies for the computational part, complemented with state-of-the-art memories and interconnect-technologies.
Easics embedded AI solutions is an entirely different market segment. It focuses on local, real-time and low-latency processing inside the machine. Our solution works on-premise (close to the sensors) and in a small footprint at low cost and low power consumption. Within this embedded space, we will focus on applications benefiting from image recognition.
Easics will target the following markets for its AI solutions: industry 4.0, agriculture, healthcare, automotive and space.
Easics deep learning platform is based on convolutional neural networks. This framework is currently running on FPGA, and we want to explore ASIC targets as well.
Deep learning is popular in image processing and computer vision for:
- object classification: what
- object detection: what & where
- object tracking: object detection over a time series
You can start your deep learning journey here, with the following steps:
- Step 1: gather data and label them
- Step 2: select a neural network or model
- Step 3: train your model
- Step 4: embed the inference engine in your application
Easics has built a prototype AI platform for object recognition and detection using a deep learning inference engine on FPGA
Easics proposes different system approaches to embed deep learning in your product or application:
- IP core on FPGA or CPU
- IP Core on System on Module (SoM)
- Deep learning in a box, connect your machine via ethernet and make it Artificial Intelligent.
Easics deep learning solution is fast, friendly and flexible and has the following benefits for you:
We match performance with cost
We implement your interfaces e.g ethernet, PCI-express, ...
We integrate with your existing hardware or new developments
A modular solution with easy deployment
Performance: high bandwidth and low latency
Flexible: programmable accelerators and hardware images on different models
Future-proof software stack ownership,
adaptable to the fast-moving AI space and evolving model types
Turnkey deployment of convolutional neural networks (CNNs)
Long-term support and scalability
Applications that can benefit from embedded AI can be found in industry 4.0, agriculture, healthcare, automotive and space:
- Application-specific machine vision systems
- Smart cameras
- Industrial machines
- Intelligent traffic systems
- Quality control in Semiconductor, chemical or pharmaceutical production