May 11 2015
Altera Corporation will demonstrate its FPGAs and software development tools for intelligent vision-related systems design at the Embedded Vision Summit on May 12, at the Santa Clara Convention Center. The event is organized by the Embedded Vision Alliance, a leading consortium in the field of computer-vision, and it features presentations by Altera and other technology leaders, in the areas of computer vision applications, technologies, techniques, and market trends.
Altera will feature the following Altera demonstrations and solutions:
- Automotive Forward-Looking Stereo Camera System, a demonstration which features an HD quality stereo vision IP suite and object detection from partners Tokyo Tech Saneyoshi Lab and Fujisoft
- Multi-Channel High-Definition Video Analytics Platform for Smart City and Automotive applications, a demonstration that is based on Altera's Cyclone® V SoC and Enpirion® PowerSoC devices combined with partner Eutecus' MVE™ video and fusion analytics technology
- A demonstration of Lucas Kanade Dense Optical Flow, which is developed with Altera SDK for OpenCL
In addition, Deshanand Singh, Director of Software Engineering, Altera, will speak at the event at 2:30 pm, May 12, on the topic, "Efficient Implementation of Convolutional Neural Networks using OpenCL on FPGAs."
- Singh will share how convolutional neural network (CNN) algorithms, which are becoming increasingly popular in embedded applications such as vision processing, can be expressed in OpenCL and compiled directly to FPGA hardware, offering detail on code optimizations and provide comparisons with the efficiency of hand-coded implementations.
On Wednesday, May 13th, attendees can join more specific workshops, including a hands-on Altera session, "Altera SDK for OpenCL for Embedded Vision Applications on FPGAs." It will be held in the Great America Meeting Room #1, at the Santa Clara Convention Center.
To learn more about Altera's role in embedded vision visit the embedded vision applications pages on Altera.com. More information on the Embedded Vision Alliance can be found at http://www.Embedded-Vision.com.