Advancing Cricket Broadcast with Semi-Automated Highlight Generation

Revolutionizing cricket broadcasting through semi-automated highlight generation using innovative algorithms to detect bowler run-up sequences. The project aims to enhance the viewing experience by extracting full match footage and metadata, implementing cutting-edge technologies, and ensuring high-quality system performance.

• Cricket
• Broadcast
• Highlight Generation
• Algorithms
• Technology

valenzuela_e Follow

Uploaded on Sep 13, 2024 | 0 Views

Download Presentation

Please find below an Image/Link to download the presentation.

The content on the website is provided AS IS for your information and personal use only. It may not be sold, licensed, or shared on other websites without obtaining consent from the author. Download presentation by click this link. If you encounter any issues during the download, it is possible that the publisher has removed the file from their server.

Presentation Transcript

1. Semi-Automated Cricket Broadcast Highlight Generation Sahil A. K. Ramlogan, Dr Akash Pooransingh University of the West Indies, St Augustine, Trinidad and Tobago IConETech-2020, Faculty of Engineering, The UWI, St. Augustine, Trinidad and Tobago

2. Background Billion Dollar Industry Importance of Player Statistics and Analysis Careful Review of Player Footage Less than 10% of Broadcast Information Necessary IConETech-2020, Faculty of Engineering, The UWI, St. Augustine, Trinidad and Tobago

3. Objectives Extract Full Cricket Match Footage Extract Full Metadata (Text Commentary)] Investigate and Implement at least 2 Algorithms to Detect the Bowler Run-up Sequence (BRS) Merge Metadata and Output BRS Extraction Compare and Assess Quality and Accuracy of Implementation Assess Final System Performance IConETech-2020, Faculty of Engineering, The UWI, St. Augustine, Trinidad and Tobago

4. Design/Implementation Commentary Extraction Component (Python) BRS Identification and Extraction (Python) Database Storage, Search and Video Playback (C# and MySQL) IConETech-2020, Faculty of Engineering, The UWI, St. Augustine, Trinidad and Tobago

5. Commentary Extraction Metadata From Cricket Websites eg: Cricbuzz.com Stored into Text Files within Specified Folder Match or Series Naming Convention IConETech-2020, Faculty of Engineering, The UWI, St. Augustine, Trinidad and Tobago

6. BRS Identification & Extraction Footage from One Innings 6 Training Frames (Covering Each BRS Scenario, Retrain per Innings) Key Point Matching Using ORB (Frame by Frame Analysis) 7 Consecutive BRS Frame Matches Constitute a Candidate BRS Validation (No or Yes) BRS Recording and Storage (Video Names) IConETech-2020, Faculty of Engineering, The UWI, St. Augustine, Trinidad and Tobago

7. Database Storage Import Videos and Commentary Text File Patterns in the Commentary Extraction of Metadata from Commentary File (Bowler, Batsman, etc.) Each Database Record/Line of Commentary Corresponds to a BRS Video Commentary and Video are in Chronological Order IConETech-2020, Faculty of Engineering, The UWI, St. Augustine, Trinidad and Tobago

8. Database Search & Video Playback Clickable Column for Video Playback Search Using (Bowler Name etc.) Construct and Execute Select Statement IConETech-2020, Faculty of Engineering, The UWI, St. Augustine, Trinidad and Tobago

9. Results Figure 1: An Example of Extracted Commentary IConETech-2020, Faculty of Engineering, The UWI, St. Augustine, Trinidad and Tobago

10. Results Figure 2: ORB Key Point Analysis and Comparison IConETech-2020, Faculty of Engineering, The UWI, St. Augustine, Trinidad and Tobago

11. Results Figure 3: Extracted BRS Video Files IConETech-2020, Faculty of Engineering, The UWI, St. Augustine, Trinidad and Tobago

12. Results Table 1: BRS Identification Statistics 1 Frame 2 Frames 6 Frames 6 Frames - Refinement 6 Frames (Verification) 1st Innings 121 6 Frames (Verification) 2nd Innings Correct Matches False Matches Deliveries Missed Matches Found Accuracy 58 81 120 121 119 32 54 75 62 0 0 66 43 4 3 3 2 86 135 195 183 121 119 46.8% 65.3% 96.8% 97.5% 97.5% 98.3% False Match Rate 37% 40% 38% 33.9% 0% 0% IConETech-2020, Faculty of Engineering, The UWI, St. Augustine, Trinidad and Tobago

13. Results Figure 4: Database with Stored Commentary Information and Video Files IConETech-2020, Faculty of Engineering, The UWI, St. Augustine, Trinidad and Tobago

14. Problems Encountered/Limitations Highlight Transitions Between Deliveries Late Transitions Lead to Missed BRS Problems with the Commentary and Video Synchronisation IConETech-2020, Faculty of Engineering, The UWI, St. Augustine, Trinidad and Tobago

15. Future Work Accounting for Missed BRS Segments Parallel Processing to Increase the Speed of the Algorithm IConETech-2020, Faculty of Engineering, The UWI, St. Augustine, Trinidad and Tobago

16. Future Work Figure 5: Difference in Time Between Each Detected Delivery IConETech-2020, Faculty of Engineering, The UWI, St. Augustine, Trinidad and Tobago

17. References Fraser Carson, Utilizing video to facilitate reflective practice: Developing sports coaches, International Journal of Sports Science & Coaching, vol. 3, no. 3, pp. 381 390, 2008. Yong Rui, Anoop Gupta, and Alex Acero, Automatically extracting highlights for tv baseball programs, in Proceedings of the Eighth ACM International Conference on Multimedia, New York, NY, USA, 2000, MULTIMEDIA 00, pp. 105 115, ACM. D. Ringis and A. Pooransingh, Automated highlight generation from cricket broadcasts using orb, in 2015 IEEE Pacific Rim Conference on Communications, Computers and Signal Processing (PACRIM), Aug 2015, pp. 58 63. IConETech-2020, Faculty of Engineering, The UWI, St. Augustine, Trinidad and Tobago

18. THANK YOU! IConETech-2020, Faculty of Engineering, The UWI, St. Augustine, Trinidad and Tobago