The Embedded Vision Alliance’s first in-person, hands-on technical training classes for 2018, Deep Learning for Computer Vision with TensorFlow, take place in just a few weeks in San Jose, California. Sponsored by IBM and Nimbix and leveraging IBM PowerAI on the Nimbix cloud platform, the classes give you the critical knowledge you need to develop deep learning computer vision applications with TensorFlow. The one-day class is on January 29 and the three-day intensive runs from January 30 to February 1. Details, along with dates for upcoming classes in Seattle, New York, Boston, and Santa Clara, can be found here.
Are you an early-stage start-up company developing a new product or service incorporating or enabling computer vision? Do you want to raise awareness of your company and products with vision industry experts, investors and entrepreneurs? Want a chance to win $5,000 in cash plus membership in the Embedded Vision Alliance? If so, apply for a chance to compete in the Vision Tank, part of the Embedded Vision Summit, which will take place May 22-24, 2018 in Santa Clara, California. The Vision Tank is the Embedded Vision Summit's annual start-up competition, showcasing the best new ventures using computer vision in their products or services. Also, register to attend the Embedded Vision Summit while Super Early Bird discount rates are still available, using discount code NLEVI0116.
Editor-In-Chief, Embedded Vision Alliance
Positions Available for Computer Vision Engineers at DEKA Research
Legendary inventor Dean Kamen founded DEKA to focus on medical innovations aimed to improve lives around the world. DEKA's team of over five hundred professionals apply their engineering, design, manufacturing and quality expertise to make DEKA a hot spot for creating innovative solutions and advanced technologies. DEKA is one of the leading research and development companies in the country and is the birthplace of some of the most innovative and life-changing products of our time. DEKA now seeks engineers with expertise in all facets of practical computer vision, from algorithm development to technology selection to system integration and testing. Interested candidates should send a resume and cover letter to firstname.lastname@example.org.
VISION PROCESSING AT THE EDGE
Edge Intelligence: Visual Reinforcement Learning for Mobile Devices
Real-life visual data encompasses a tremendous amount of information and presents a huge challenge for the design and development of a perceptual engine. Smart machines equipped with visual understanding technology will always be presented and challenged with new data. In this talk, Adham Ghazali, co-founder and CEO of Imagry, presents algorithm methods to allow learning in the end device to enable it to understand new data. The main challenges to learning in the end device stem from the lack of computing power, lack of access to sufficient data samples and the need for involvement of human experts. Imagry addresses these challenges by combining binary weights representation of deep neural networks and reinforcement learning. In particular, Ghazali explores and introduces a self-expanding cost function and the incorporation of external memory to enable DNNs to adapt to new data.
Computer Vision and Machine Learning at the Edge
Computer vision and machine learning techniques are applied to myriad use cases in smartphones today. As mobile technology expands beyond the smartphone vertical, both technologies will continue to fuel innovation, individually and in concert. In this presentation, Michael Mangan, a member of the Product Manager Staff at Qualcomm Technologies, discusses Qualcomm Technologies, Inc.’s use of and vision for the future of computer vision and machine learning at the edge.
DEEP LEARNING APPLICATIONS
End to End Fire Detection Deep Neural Network Platform
This presentation from Divya Jain, Technical Director at Tyco Innovation, dives deep into a real-world problem of fire detection to see what it takes to build a complete solution using CNNs. Fire is specifically challenging because it doesn’t have a fixed shape or size like other objects. The presentation begins with a discussion of the technology stack, followed by the algorithm, and concluding with a review of the end to end architecture. Jain discusses the challenges her company encountered while training this algorithm and how they worked through them by building a scalable and reusable platform.
Deep Learning Beyond Cats and Cars: Developing a Real-life DNN-based Embedded Vision Product for Agriculture, Construction, Medical, or Retail
By now we know very well how to design and train a neural network to recognize cats, dogs and cars. But what about real projects — for example, in agriculture, construction, medical, and retail? This how-to talk from Alexey Rybakov, Senior Director for Embedded Systems at Luxoft, provides an overview of what it takes to design, train, and fine-tune a real-life DNN-based embedded vision solution. Rybakov explores algorithmic, data set, training, and optimization decisions that take you from proofs-of-concepts to solid, reliable, and highly optimized systems. This material is based on Luxoft's own successes, failures, and lessons learned while implementing embedded vision solutions.