Automatic Curation of Golf Highlights using Multimodal Excitement Features
This addresses the problem of reducing manual video editing for broadcast media in sports, specifically golf, though it appears incremental as it builds on existing multimodal analysis techniques.
The paper tackles the labor-intensive task of producing sports highlight packages by proposing a novel approach for auto-curating golf highlights, fusing multimodal excitement features from players, spectators, and commentators to identify key moments with metadata like player names and hole numbers, and it was successfully demonstrated at a major tournament over four days.
The production of sports highlight packages summarizing a game's most exciting moments is an essential task for broadcast media. Yet, it requires labor-intensive video editing. We propose a novel approach for auto-curating sports highlights, and use it to create a real-world system for the editorial aid of golf highlight reels. Our method fuses information from the players' reactions (action recognition such as high-fives and fist pumps), spectators (crowd cheering), and commentator (tone of the voice and word analysis) to determine the most interesting moments of a game. We accurately identify the start and end frames of key shot highlights with additional metadata, such as the player's name and the hole number, allowing personalized content summarization and retrieval. In addition, we introduce new techniques for learning our classifiers with reduced manual training data annotation by exploiting the correlation of different modalities. Our work has been demonstrated at a major golf tournament, successfully extracting highlights from live video streams over four consecutive days.