Indian researchers from the International Institute of Information Technology, Hyderabad (IIIT Hyderabad) have used machine-learning techniques to generate text-based cricket commentary with an accuracy rate of 90%.
The video is segmented into "scenes" using scene category information extracted from text commentary. The system then classifies video shots and the phrases in the textual description into various categories. Finally, the relevant phrases are mapped into the video shots.
The researchers say the technique could be used by sports websites to automate and assist reporters in writing real-time cricket commentary.
The video dataset was collected from the Indian Premier League tournament's YouTube channel, while small samples of commentary were taken from commentary of about 300 matches on Cricinfo.
The researchers note the algorithms were able to accurately label a batsman's cricketing shot by using visual-recognition techniques on an action that lasts just 1.2 seconds.
Annotation of the videos enables the researchers to build a retrieval system that can search across hundreds of hours of content for specific actions that last only a few seconds.
The researchers believe cricket teams could use the technology to analyze strengths and weaknesses of particular players, and it also could be applied to other sports such as soccer or tennis.
From NDTV (India)
View Full Article
Abstracts Copyright © 2015 Information Inc., Bethesda, Maryland, USA