Raspberry Pi 3 and Movidius Neural Compute Stick to the Rescue Against Child Pornography

Raspberry Pi Powered Nudity Filter

Warning note: While there won’t be any NSFW photos in this post, there will be some photos of ladies in light clothing (e.g. bikini) and “naked” animals for testing purpose… Intel released Movidius Neural Compute Stick allowing low power image recognition at the edge earlier this year, and we’ve seen it work just fine with Raspberry Pi 3 board delivering three times the performance against an inference solution leveraging VideoCore IV GPU. Christian Haschek owns a photo hosting site (PictShare) which happens to run open source code with the same name, and allows user to upload images anonymously. However, he soon found out that at least one user uploaded some child pornography. He contacted the authorities, but then wondered whether there may be others? Since there are simply too many photo on the website to look it up manually, he decided to look for a solution, and went with a few Raspberry Pi 3 boards, some Neural Compute Sticks, as …

Imagination Unveils PowerVR AX2185 and AX2145 Neural Network Accelerators (NNA)

PowerVR AX2185

Imagination Technologies introduced PowerVR Series2NX neural network accelerator (NNA) last year. At the time, the company claimed the NNA would deliver  twice the performance at half the bandwidth over existing competing solutions, and that it was the first dedicated hardware solution with flexible bit-depth support from 16-bit down to 4-bit. What they did not announce last September were any specific parts. But they’ve just done that today with PowerVR AX2185 and AX2145 NNAs “designed to enable high-performance computation of neural networks at very low power consumption in minimal silicon area”. PowerVR AX2185 PowerVR AX2185 is said to be the highest performing neural network accelerator per mm2 in the market with eight full-width compute engines delivering up to 4.1 TOPS (Tera Operations Per Second), or 2,048 MACs/clock. AX2185 NNA can deliver 3.5 times the performance compared to a desktop GPU. The NNA is best suited for high-end smartphones, smart surveillance and automotive markets. PowerVR AX2145 PowerVR AX2145 is a lower cost …

NVIDIA Introduces Jetson Xavier Devkit and Isaac Robotics Software

NVIDIA Jetson Xavier

NVIDIA Xavier was first unveiled in September 2016 as an artificial intelligence SoC with eight NVIDIA Custom 64-bit Arm cores, a 512-core Volta GPU,  8K video encoding and decoding, and a computer vision accelerator (CVA) now called NVDLA (NVIDIA Deep Learning Accelerator). Earlier this year, the company announced Xavier was sampling,  and DRIVE IX & DRIVE AR SDKs for the automotive market. On the eve of Computer 2018, NVIDIA has introduced Jetson Xavier development kit, as well as Isaac robotics software for autonomous machines. Jetson Xavier key specifications: SoC – NVIDIA Xavier with 8-core ARMv8.2 64-bit CPU, 8MB L2 + 4MB L3 512-core Volta GPU with Tensor Cores 2x NVDLA engines for deep learning 7-way VLIW Processor for vision acceleration VPU with dual 4Kp60 video decoding and encoding System Memory – 16GB 256-bit LPDDR4x | 137 GB/s Storage – 32GB eMMC 5.1 flash Display – 3x eDP/DP/HDMI at 4Kp60 | HDMI 2.0, DP HBR3 Camera 16x CSI-2 Lanes (40 Gbps …

96Boards Unveils Four A.I. Developer Platforms: HiKey 970, Ultra96, ROCK960 PRO & Enterprise Edition

Hikey-970

Many new processors include a Neural Processing Unit (NPU) – aka Neural Network Accelerator (NNA) – in order to speed up talks associated with artificial intelligence, such as object or other patterns recognitions. With Linaro Connect Hong Kong 2018, 96Boards has just unveiled four development boards specifically designed for artificial intelligence solution with Hikey 970 powered by Hisilicon Kirin 970 processor, Ultra96 based on Xilinx Zynq UltraScale+ ZU3EG ARM+ FPGA SoC,  and ROCK960 PRO & Enterprise Edition featuring the upcoming Rockchip RK3399Pro processor. Hikey 970 Preliminary specifications: SoC – Kirin 970 with 4x Cortex A73 @ 2.36GHz,  4x Cortex A53 @ 1.8GHz, Arm Mali G72-MP12 GPU, NPU with 256MAC/cycle @ 960MHz System Memory – 6GB 1866MHz, 4 Channel LPDDR4x Storage  -64GB UFS storage, micro SD card slot Video Output – HDMI 2.0a Camera – 4 lanes CSI + 2 lanes CSI Connectivity – Gigabit Ethernet, wireless module, GPS, GLONASS, and BeiDou USB – 2x USB 3.0 ports,  2x USB type …

Rockchip RK3399Pro SoC Integrates a 2.4 TOPS Neural Network Processing Unit for Artificial Intelligence Applications

Rockchip RK3399 (aka OP1) SoC was launched in 2016 with an hexa core Arm Cortex A72/A53 processor, Mali-T860MP4 GPU, support for 4K video decoding, and high speed interfaces like USB 3.0 and PCIe, as well as Gigabit Ethernet. The processor is found in Chromebooks, TV boxes, development boards, and other devices. The company has unveiled an upgraded “Pro” version of the processor at CES 2018. Rockchip RK3399Pro appears to have most of the same features as its predecessor but adds a neural network processing unit (NPU) delivering up to 2.4 TOPS for artificial intelligence and deep learning applications. The company claims that compared to traditional solution, the computing performance of typical deep neural network Inception V3, ResNet34 and VGG16 models on RK3399Pro is improved by almost one hundred times, and power consumption is less than 1% than A.I. solutions implemented using GPU acceleration. Based on the information provided in the chart above (source: elDEE on twitter), Rockchip RK3399Pro outperforms other …

Laceli AI Compute Stick is a More Powerful & Efficient Alternative to Intel/Movidius Neural Compute Stick

Intel’s Movidius Neural Compute Stick is a low power deep learning inference kit and “self-contained” artificial intelligence (A.I.) accelerator that connects to the USB port of computers or development boards like Raspberry Pi 3, delivering three times more performance than a solution accelerated with VideoCore IV GPU. So far it was the only A.I USB stick solution that I heard of, but Gyrfalcon Technology , a US startup funded at the beginning of last year, has developed its own “artificial intelligence processor” with Lightspeeur 2801S, as well as a neural USB compute stick featuring the solution: Laceli AI Compute Stick. The company claims Laceli AI Compute Stick runs at 2.8 TOPS (Trillion operation per second) performance within 0.3 Watt of power, which is 90 times more efficient than the Movidius USB Stick that can deliver 100 GFLOPS (0.1 TOPS) within 1 Watt of power. Information about the processor and stick is rather limited, but Gyrfalcon explains their APiM architecture (AI …

Khronos Group Releases Neural Network Exchange Format (NNEF) 1.0 Specification

The Khronos Group, the organization behind widely used standards for graphics, parallel computing, or vision processing such as OpenGL, Vulkan, or OpenCL, has recently published NNEF 1.0 (Neural Network Exchange Format) provisional specification for universal exchange of trained neural networks between training frameworks and inference engines. NNEF aims to reduce machine learning deployment fragmentation by enabling data scientists and engineers to easily transfer trained networks from their chosen training framework into various inference engines via a single standardized exchange format. NNEF encapsulates a complete description of the structure, operations and parameters of a trained neural network, independent of the training tools used to produce it and the inference engine used to execute it. The new format has already been tested with tools such as TensorFlow, Caffe2, Theano, Chainer, MXNet, and PyTorch. Khronos has also released open source tools to manipulate NNEF files, including a NNEF syntax parser/validator, and example exporters, which can be found on NNEF Tools github repository. The provisional NNEF …

Qualcomm Developer’s Guide to Artificial Intelligence (AI)

Qualcomm has many terms like ML (Machine Learning), DL (Deep Learning), CNN (Convolutional Neural Network),  ANN (Artificial Neural Networks), etc.. and is currently made possible via frameworks such as TensorFlow, Caffe2 or ONNX (Open Neural Network Exchange). If you have not looked into details, all those terms may be confusions, so Qualcomm Developer Network has released a 9-page e-Book entitled “A Developer’s Guide to Artificial Intelligence (AI)” that gives an overview of all the terms, what they mean, and how they differ. For example, they explain that a key difference between Machine Learning and Deep Learning is that with ML, the input features of the CNN are determined by humans, while DL requires less human intervention. The book also covers that AI is moving to the edge / on-device for low latency, and better reliability, instead of relying on the cloud. It also quickly go through the workflow using Snapdragon NPE SDK with a total of 4 steps including 3 …