InsideDSP — In-depth analysis and opinion

Jeff Bier’s Impulse Response—Why SLAM is Becoming the New GPS

GPS has proven to be an extraordinarily valuable and versatile technology. Originally developed for the military, today GPS is used in a vast and diverse range of applications. Millions of people use it daily for navigation and fitness. Farmers use it to manage their crops. Drones use it to automatically return to their starting location. Railroads use it to track train cars and other equipment. In the 40 years since the first GPS satellite was launched, GPS receivers have shrunk dramatically Read more...

Case Study: Using Deep Learning Techniques for Commercial Product Success

Here at BDTI, we’re working to apply deep learning techniques to a wide range of applications. Deep learning can be extremely effective—if there’s the right combination of data and processing power. The challenge to success lies in understanding how much data is sufficient and how to process it efficiently. This is where BDTI’s expertise in algorithms and architectures delivers value to our customers. Recently, BDTI was engaged to create a convolutional neural network (CNN) to classify items Read more...

CEVA's XC12 DSP Core Targets Multi-Gbit Cellular, Wi-Fi Opportunities

At last month's MWC (Mobile World Congress), CEVA unveiled the XC12, its latest DSP core for communication infrastructure applications. A significant feature set upgrade to the XC4500 DSP core precursor, the XC12 is tailored for the bandwidth, latency and user count demands of next-generation "5G" cellular-supportive infrastructure equipment, ranging from remote digital front-end radio heads through multi-mode baseband processing in various-sized base stations, and all the way to wireless Read more...

NVIDIA's Jetson TX2: Embedded Designs Gain a Deep Learning Upgrade

Jetson TX2 is NVIDIA's latest board-level product targeted at computer vision, deep learning, and other embedded AI tasks, particularly focused on "at the edge" inference (when a neural network analyzes new data it’s presented with, based on its previous training) (Figure 1). It acts as an upgrade to both the Tegra K1 SoC-based Jetson TK1, covered in InsideDSP in the spring of 2014, and the successor Tegra X1-based Jetson TX1, which BDTI evaluated for deep learning and other computer vision Read more...

Jeff Bier’s Impulse Response—Will Deep Learning Displace All Other Computer Vision Techniques?

It's remarkable to see the range of applications in which deep neural networks are proving effective – often, significantly more effective than previously known techniques. From speech recognition to ranking web search results to object recognition, each day brings a new product or published paper with a new challenge tamed by deep learning. Computer vision, of course, is a field with significant deep learning activity. Deep learning is particularly appealing for visual perception because Read more...

Case Study: Hands-On Experience Delivers Insights Into Technologies and Trends for Deep Learning

One of the most interesting resources for those interested in the rapidly growing field of deep learning is the "Cognitive Computing Startup List," compiled by Chris Rowan of Cognite Ventures. As of February 21, Chris had identified 275 companies that are, in his words, "the most focused, the most active and the most innovative…." These were culled from various lists of companies that tout artificial intelligence and deep learning. Such companies number in the thousands—the sheer number is a Read more...

Himax, CEVA, emza Partner to Develop Low Power Vision Processing Platform

Image-sensor maker Himax, processor core vendor CEVA, and algorithm provider emza Visual Sense have partnered to develop a low-power "always on" vision sensor module with integrated visual analytics. The three companies offered a conceptual demonstration of the product, dubbed the WiseEye IoT vision sensor, at last month's Consumer Electronics Show. Assuming that WiseEye IoT is able to perform with sufficient accuracy and frame rates, with its target power consumption of 5-10 mW for common Read more...

Jeff Bier’s Impulse Response—Deployable Artificial Neural Networks Will Change Everything

In recent months, evidence has continued to mount that artificial neural networks of the "deep learning" variety are significantly better than previous techniques at a diverse range of visual understanding tasks. For example, Yannis Assael and colleagues from Oxford have demonstrated a deep learning algorithm for lip reading that is dramatically more accurate than trained human lip readers, and much more accurate than the best previously published algorithms. Meanwhile, Andre Esteva, Brett Read more...

Case Study: Deep Understanding of Processor Architectures and Computer Vision Algorithms is Key to a Breakthrough Product

Computer vision promises to be the key to the next set of "killer apps"—computer vision-enabled apps that will leverage artificial intelligence to help keep us safer and healthier. But computer vision algorithms are compute- and power-intensive, and need to process large amounts of data. These barriers have limited their use to enterprise, line-powered devices and cloud-assisted mobile devices. The implementation of computer vision algorithms on mobile processors, where compute resources are Read more...

New AMD Software Library, Hardware Support Deep Learning Acceleration

HPC (high-performance computing) servers, which have notably embraced the GPGPU (general-purpose computing on graphics processing units) concept in recent years, are increasingly being employed for computer vision and other deep learning-based applications. Beginning in late 2014, NVIDIA supplemented its general-purpose CUDA toolset for GPU-accelerated heterogeneous computing with its proprietary CuDNN software library, which codifies the basic mathematical and data operations at the core of Read more...