No
Yes
View More
View Less
Working...
Close
OK
Cancel
Confirm
System Message
Delete
Schedule
An unknown error has occurred and your request could not be completed. Please contact support.
Scheduled
Wait Listed
Personal Calendar
Speaking
Conference Event
Meeting
Interest
Schedule TBD
Conflict Found
This session is already scheduled at another time. Would you like to...
Loading...
Please enter a maximum of {0} characters.
Please enter a maximum of {0} words.
must be 50 characters or less.
must be 40 characters or less.
Session Summary
We were unable to load the map image.
This has not yet been assigned to a map.
Search Catalog
Reply
Replies ()
Search
New Post
Microblog
Microblog Thread
Post Reply
Post
Your session timed out.
This web page is not optimized for viewing on a mobile device. Visit this site in a desktop browser to access the full set of features.
2017 GTC San Jose

S7671 - Real-Time Live Video Highlight Identification at Scale: Lessons Learned from Yahoo eSports

Session Speakers
Session Description

The Yahoo eSports video highlight identification system, which is live in production, is a large-scale, real-time, cloud computing system that leverages computer vision, deep learning, and GPU technologies. We'll share the challenges we faced and our solutions for building such a high-performance, large-scale deep learning and inference system. Specifically, we'll discuss our data annotation process that drastically reduced time required for collecting large-scale data, the use of CaffeOnSpark over Yahoo's GPU clusters to get massive training time improvement, trade-offs between CPU and GPU for performing real-time inference on the cloud, and several engineering tricks that allowed us to achieve real-time performance on live video.


Additional Session Information
All
Talk
Computer Vision and Machine Vision Deep Learning and AI
Internet / Telecommunications
50 minutes
Session Schedule