Ashish Vikram
Specialises In
Ashish, co-founder and CTO of VideoKen, is a software industry veteran. A graduate of IIT Delhi and Purdue University, he started his career in the Silicon Valley in 1989. He moved to Bangalore and set up three new software development organizations from scratch, grew and nurtured them into highly productive software development labs until eventual acquisition. Prior to co-founding VideoKen he was leading engineering for the ads group at Flipkart where he built a machine learning-centered ad technology stack for brand ads and product listing ads from the ground up. Prior to Flipkart he was VP Engineering at Yahoo, Bangalore where he led the Ads and Data engineering teams for 3 years.
Ashish has extensive experience in managing technical projects/products - both in USA and India. Several years of hands-on design and development experience in systems software, strong architectural, analytical, programming and algorithmic skills. Ashish's specialty is in building and managing highly productive high-quality engineering teams for developing software products.
-
keyboard_arrow_down
Lighting up the Blackhole of the Internet using AI
20 Mins
Experience Report
Beginner
Videos account for about 75% of Internet traffic today. Enterprises are creating more and more videos and using them for various informational purposes, including marketing, training of customers, partners & employees and internal communications. However, video
s are considered as the blackholes o f the Internet because it is very hard to see what’s inside them. The opaque nature of videos equally impacts end users who spend a lot of time navigating to their point of interest, leading to severe underutilization of videos as a powerful medium of information. In this talk, we will describe visual processing pipeline of VideoKen platform which includes
- Graph-based algorithm along with deep scene text detection to identify key visual frames in the video,
- FCN-based algorithm for semantic segmentation of screen content in visual frames,
- Transfer-learning based visual classifier to categorize screen content into different categories such as slides, code walkthrough, demo, handwritten, etc. and
- Algorithm to detect visual coherency and select indices from the video.
We will discuss challenges and experiences in implementing/iterating on these algorithms using our experience with processing 100K+ video hours of content.
-
No more submissions exist.
