Subscribe Contact Us


CDNN2 - CEVA Deep Neural Network Software Framework

Accelerating the development of Artificial
Intelligence and its deployment in
Low-Power Embedded Systems

*CEVA Deep Neural Network 2nd gen
CDNN2 is a second generation neural network software framework capable of running deep learning-based video analytics directly on any camera-enabled device, in real time.
CDNN2 builds on the successful foundations of CEVA’s first generation neural network software framework, offering improved capabilities and performance for the most sophisticated and latest network topologies and layers, including support for Caffe and TensorFlow, Google’s software library for machine learning.

Coupled with the
CEVA-XM4 intelligent vision processor, CDNN2:

Offers significant time-to-market advantages for implementing machine learning in embedded systems – saves months in the deployment of DNNs onto real-life constraints.
Significantly improves on power consumption and memory bandwidth compared to a leading GPU-base system.
Supports any given layer in any network topology, on any resolution, as trained by Caffe and TensorFlow. Supported networks including; Alexnet, GoogLeNet, ResidualNet (ResNet), SegNet, VGG and Network-in-network (NIN).

“Seeing is believing” - see CDNN2 in action:

Generating embedded-ready neural networks from offline pre-trained floating-point networks take less than 10 minutes.
Want to see?

In the News

EETimes Europe
Configurable IoT processor handles it all 09 October 2016
Media and Entertainment Technologies
CEVA refines its Vision DSP by Peter McGuinness 27 September 2016
Learning how to learn: Toddlers vs. neural networks 27 September 2016

Need additional information? Click here to contact us