This challenge aims to classify TV news videos into two categories: commercials and non-commercials. Commercial blocks in news videos occupy almost 40-60% of total air time. The manual segmentation of commercials from thousands of TV news channels is time-consuming and economically infeasible. Hence, the need for machine learning-based methods arises. This challenge provides a dataset consisting of 150 hours of broadcast news videos from 5 different news channels, including 3 Indian and 2 international channels.
Given the audio-visual features extracted from each video shot, the objective is to classify each shot as a commercial or a non-commercial. This is a semantic video classification problem.
The evaluation metric for this challenge is the AUC.
https://archive.ics.uci.edu/ml/machine-learning-databases/00326/
Dr. Prithwijit Guha , Raghvendra D. Kannao and Ravishankar Soni Multimedia Analytics Lab, Department of Electrical and Electronics Engineering, Indian Institute of Technology, Guwahati, India