Deep learning has recently become an abundant technology for analyzing video data. However, the increasing resolution and frame rates of videos makes real-time analysis a remarkably challenging task. Here, we will demonstrate a novel architecture that enables real-time deep learning analysis of high resolution video streaming. Our architecture can serve advanced deep learning algorithms at sub second rates. We illustrate this specifically using a task aimed at detecting corruptions and various errors in video rendering. Our solution engulfs multiple technologies including Redis, Docker and Tensorflow, and appears fully synchronous to the user despite containing an asynchronous backend. Overall, our results highlight the feasibility of our solution in enablement of real-time neural networks processing of videos. This approach is generalizable and can be applied to diverse domains which requires video analytics.
Eran is a senior software engineer in the Advanced Analytics department at Intel where he enjoys everything distributed. From Spark and Kafka to Kubernetes and Tensorflow, He loves playing with them all. Eran poses a MS in computer science from the Hebrew University of Jerusalem.
Help us make this conference the best it can be for you. Have questions you'd like this speaker to address? Suggestions for issues that deserve extra attention? Feedback that you'd like to share with the speaker and other attendees?
Join the conversation here (requires login)
©2018, O’Reilly UK Ltd • (800) 889-8969 or (707) 827-7019 • Monday-Friday 7:30am-5pm PT • All trademarks and registered trademarks appearing on oreilly.com are the property of their respective owners. • email@example.com