
Neural network DSP core targets vision, radar/lidar, and fused-sensor apps
Cadence Design Systems (San Jose, CA) positions this as the industry’s first standalone, self-contained neural network DSP IP core optimized for vision, radar/lidar and fused-sensor applications with high-availability neural network computational needs. Targeted for the automotive, surveillance, drone and mobile/wearable markets, the Vision C5 DSP offers 1TMAC/sec computational capacity to run all neural network computational tasks.
The contrast drawn with approaches using accelerators specific to the ‘primitives’ of neural net computing, is that that approach can involve moving large amounts of data between processing elements and memory, which is power-hungry. By building an architecture that places all the layered aspects of the neural net model in one place, there are great efficiencies to be realised, Cadence asserts. C5 is, in effect, a general purpose computational machine, but shaped arount the core functions of neural net computing.
As neural networks get deeper and more complex, the computational requirements are increasing rapidly. Meanwhile, neural network architectures are changing regularly, with new networks appearing constantly and new applications and markets continuing to emerge. These trends are driving the need for a high-performance, general-purpose neural network processing solution for embedded systems that not only requires little power, but also is highly programmable for future-proof flexibility and lower risk.
Camera-based vision systems in automobiles, drones and security systems require (Cadence says) two fundamental types of vision-optimized computation. First, the input from the camera is enhanced using traditional computational photography/imaging algorithms. Second, neural-network-based recognition algorithms perform object detection and recognition.
Existing neural network accelerator solutions are hardware accelerators attached to imaging DSPs, with the neural network code split between running some network layers on the DSP and offloading convolutional layers to the accelerator. This combination is inefficient and consumes unnecessary power.
Architected as a dedicated neural-network-optimized DSP, the Vision C5 DSP accelerates all neural network computational layers (convolution, fully connected, pooling and normalization), not just the convolution functions. This frees up the main vision/imaging DSP to run image enhancement applications independently while the Vision C5 DSP runs inference tasks. By eliminating extraneous data movement between the neural network DSP and the main vision/imaging DSP, the Vision C5 DSP provides a lower power solution than competing neural network accelerators. It also offers a simple, single-processor programming model for neural networks.
“Many of our customers are in the difficult position of selecting a neural network inference platform today for a product that may not ship for a couple of years or longer,” said Steve Roddy, senior group director, Tensilica marketing at Cadence.
“Not only must neural network processors for always-on embedded systems consume low power and be fast on every image, but they should also be flexible and future proof. All of the current alternatives require undesirable tradeoffs, and it was clear a new solution is needed. We architected the Vision C5 DSP as a general-purpose neural network DSP that is easy to integrate and very flexible, while offering better power efficiency than CNN accelerators, GPUs and CPUs.”
The Vision C5 DSP is a self-contained engine with;
– 1TMAC/sec computational capacity (4X greater throughput than the Tensilica Vision P6 DSP) for very high computation throughput on deep learning kernels
– 1024 8-bit MACs or 512 16-bit MACs for performance at both 8-bit and 16-bit resolutions
– VLIW SIMD architecture with 128-way, 8-bit SIMD or 64-way, 16-bit SIMD
– Architected for multi-core designs, enabling a multi-teraMAC solution in a small footprint
– Integrated iDMA and AXI4 interface
– Uses the same proven software toolset as the Vision P5 and P6 DSPs
Cadence claims that, compared to commercially available GPUs, the Vision C5 DSP is up to 6X faster in the “AlexNet” CNN performance benchmark and up to 9X faster in the “Inception V3” CNN performance benchmark. The C5 DSP supports variable kernel sizes, depths and input dimensions. It also accommodates several different coefficient compression/decompression techniques, and support for new layers can be added as they evolve.
The Vision C5 DSP also comes with the Cadence neural network mapper toolset, which will map any neural network trained with tools such as Caffe and TensorFlow into executable and optimized code for the Vision C5 DSP, making use of a set of hand-optimized neural network library functions.
Cadence: www.cadence.com/go/visionc5
