System and software developers who want to build their skills in practical computer vision implementation techniques will want to check out the in-depth workshops being presented on May 11th and 13th in Santa Clara, California, in conjunction with the May 12th Embedded Vision Summit. Details on all four of the currently planned workshops are on the Alliance website. I'll discuss two of the workshops in detail here; a prior article covered the other two workshops, one from Altera and the other from BDTI and Khronos.
"Enabling Computer Vision on ARM" provides expert guidance on implementing computer vision applications on ARM platforms. In this workshop, a select group of computer vision experts and leaders will present their experiences working with ARM-based systems across a variety of real use cases. Planned presentations include talks on the following topics:
- Benchmarking Metrics and Processor Selection for Computer Vision
- Implementing Real-Time 3D Object Recognition on Mobile Devices
- Accelerated Computer Vision Processing in Augmented Reality
- Holistic Profiling of the CPU and GPU for Optimal Vision Applications
- An OpenCL Case Study on Real-time Dense Passive Stereo Vision, and
- Image and Vision Processing Software Development and Optimizations
The workshop will also include an industry panel, in which participants from the worlds of vision IP, ADAS, and image sensors and recognition software will discuss future trends in technology for ARM-based systems. "Enabling Computer Vision on ARM" takes place on May 11th from 1 PM to 5:30 PM, and attendance is free. Register today, as capacity is limited.
Synopsys' workshop, "Using the DesignWare Embedded Vision Processor for Low-Power, Low-Cost Video Surveillance and Object Detection Applications," also takes place on May 11, from 12 PM to 5:30 PM, and attendance is also free. In it, you will discover the features of the new Synopsys’ DesignWare EV5x Embedded Vision Processor and its application to video surveillance and object detection applications. The workshop will present a top-down capture and mapping of representative vision applications using an OpenVX programming environment. And it will describe the EV5x processor, which Synopsys designed to accelerate convolutional neural networks (CNNs), a topic that will be covered extensively at the Embedded Vision Summit. Online registration is now available; again, promptness on your part is strongly encouraged.
Complementing these workshops, the Embedded Vision Summit agenda includes numerous business and technical presentations by industry leaders, along with exciting keynote talks and a state-of-the-art technology showcase. Last year's Summit sold out, so don't delay: register today!
