Automatic summarization of broadcast cricket videos

Abstract

Summarization of cricket videos is very important because of three reasons: 1) its long duration making manual highlights generation tedious 2) less explored area compared to other sports like soccer 3) huge viewership. We propose a novel summarization scheme for cricket which exploits its contextual semantics. First, we detect the bowling frames based on which the video is temporally segmented into individual deliveries. Then each temporal segment representing a delivery is classified into an interesting or non-interesting segment based on detection of events namely boundaries and wickets. Due to the high frequency of ads and replays in cricket, we have proposed robust algorithms for their removal. Finally, we have proposed a finite state automaton based modeling of the temporal segments to extract key-frames. We have also extended the framework to include text cues and expert choices and also developed a hierarchical summary. We have tested our algorithm on several broadcast cricket videos and obtained good results.

CricShotClassify: An Approach to Classifying Batting Shots from Cricket Videos Using a Convolutional Neural Network and Gated Recurrent Unit

“…Some studies were conducted for automatic highlight extraction from cricket videos [ 14 , 15 , 16 , 17 , 18 , 19 ]. In [ 14 ], video and audio features were combined.…”

“…A color histogram (CH) and a histogram of oriented gradients (HOG) were used in their work. A novel summarization scheme was proposed in [ 18 ]. In this work, boundary and wicket events were extracted on the basis of bowling and score change.…”

Broadcast Cricket Highlight Treatment using a Quadrant Based Method for Global Motion

“…Results were aligned with text based commentary to determine highlights. Kumar et al in [18] detected the 'bowling frames' but provided a higher level of semantic access as 'bowling frames' were further classified into an interesting or noninteresting segment based on detection of events (boundaries and wickets). This method used the k-means algorithm to segment frames and utilized these results for a zoom-in detection based on the width of the bounding box of the pitch.…”