Tutorial video: object detection using HOG and SVM
At the Embedded Vision Summit in May, our Marco Jacobs presented the tutorial “Implementing Histogram of Oriented Gradients on a Parallel Vision Processor”. Object detection in images is one of the core problems in computer vision. The Histogram of Oriented Gradients (HOG) method and Support Vector Machines (SVM) are key algorithms for this task. They’re commonly used in automotive and surveillance applications such as pedestrian detection or perimeter protection.
In this presentation, Marco gives a high-level overview of the algorithm and shows how it can be implemented in real-time on our high-performance, low-cost, and low-power v-MP4000HDX processor architecture.
Show report: Image Sensors Auto conference
The Image Sensors conferences have been held since 2008, and have grown to become one of the main image sensor conferences in Europe. This was the first image sensor conference from this organization that focused entirely on automotive applications. The two-day conference in June attracted about one hundred attendees, each with key roles in the automotive supply chain. Everyone working with cameras for automotive seemed to be there: semiconductor IP companies, image sensor makers, automotive SOC vendors, Tier 1s, and the OEMs. There were sixteen presentations, addressing a wide range of topics. Not only covering image sensors, but also the other components that make up an imaging system for automotive applications.
The following topics were recurring themes at the conference that were most often discussed: ADAS, viewing applications, image quality, safety, thermal design and sensor fusion.
Employee interview: Tony Picard, VP Sales
This is a new section in our newsletter, and we’re kicking it off together with our VP Sales, Tony Picard.
Why did you decide to join videantis?
Videantis provides an ideal combination of a strong market potential with key technology that addresses it. Cameras are growing explosively, both in consumer and automotive, at twice the rate of the semiconductor industry. The latest Amazon Firefly mobile phone has six cameras for instance, and the latest BMW X5 has five. In addition, there’s a clear market trend of combining computer vision with video coding. Whenever you have a camera, and you want to store or transmit these images, you need to compress the data, so you need codecs. Making all of these cameras smart is another trend, allowing new user experiences on our mobile phones, and making our cars 100% safe and more comfortable to drive.
Interesting industry news
Amazon Fire phone includes six cameras, using four IR cameras for computer vision
The Verge, Mashable: The Amazon Fire Phone uses four front-facing IR cameras to create a 3D user interface. “The question will be whether or not the extra processing power and potential battery life drain to run these multiple cameras will be worth it.” You really need a small vision DSP such as ours to really make this low power.
GoPro’s IPO shows video-centric products thrive
On June 25th, GoPro went public, valuing the company at over $5B as of July 10th. Go to GoPro’s website and you’ll see what they’re all about: capturing high quality video in a low-power mobile package, offering it at consumer price points. We’re excited about this IPO since they’re all about video, and they’re doing very well. Just like us. We wish them all the best!
The self-driving car will make driving much cheaper
In 5 to 7 years, it’ll cost us more to drive our cars than to let them drive us. There are two types of cost to consider: the cost of accidents and the cost of operation. In the U.S. alone, vehicular accidents inflict about $450 billion in annual costs. With respect to operation, cars are used less than 5% of the time. Autonomous-car sharing will be a much more convenient and cost-efficient mode of transportation for many, reducing cost of ownership by an additional 2/3rds. Read the full article.
|IBC||Sep 13-17, 2014, Amsterdam, the Netherlands||Talk to us at this global broadcast event|
|ARM TechCon||October 1-3, 2014, Santa Clara, California||Meet us at this key SOC design event|
|2014 Ethernet & IP @ Automotive Technology Day||23-24 October 2014, Detroit, Michigan||Come see our automotive demonstrations|
Schedule a meeting with us by sending an e-mail to firstname.lastname@example.org. We’re always interested in discussing your video and vision SOC design ideas and challenges. We look forward to talking with you!