Computational sports broadcasting: Automated director assistance for live sports

Chen, Christine; Wang, Oliver; Heinzle, Simon; Carr, Peter; Smolic, Aljoscha; Groß, Markus

doi:10.1109/icme.2013.6607445

Cited by 25 publications

(18 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This result suggests that heuristic techniques using ball and player locations are not enough for dynamic PTZ camera selection. ADA [6] seems to have substantial challenges on this dataset. It is partially because of hand-crafted features (such as player flow) are quite noisy from fast-moving PTZ cameras.…”

Section: Discussionmentioning

confidence: 99%

Learning Sports Camera Selection From Internet Videos

Chen

Tian

et al. 2019

2019 IEEE Winter Conference on Applications of Computer Vision (WACV)

View full text Add to dashboard Cite

This work addresses camera selection, the task of predicting which camera should be "on air" from multiple candidate cameras for soccer broadcast. The task is challenging because of the scarcity of learning data with all candidate views. Meanwhile, broadcast videos are freely available on the Internet (e.g. Youtube). However, these videos only record the selected camera views, omitting the other candidate views. To overcome this problem, we first introduce a random survival forest (RSF) method to impute the incomplete data effectively. Then, we propose a spatialappearance heatmap to describe foreground objects (e.g. players and balls) in an image. To evaluate the performance of our system, we collect the largest-ever dataset for soccer broadcasting camera selection. It has one main game which has all candidate views and twelve auxiliary games which only have the broadcast view. Our method significantly outperforms state-of-the-art methods on this challenging dataset. Further analysis suggests that the improvement in performance is indeed from the extra information from auxiliary games.

show abstract

Section: Discussionmentioning

confidence: 99%

Learning Sports Camera Selection From Internet Videos

Chen

Tian

et al. 2019

2019 IEEE Winter Conference on Applications of Computer Vision (WACV)

View full text Add to dashboard Cite

show abstract

“…Recently, purely data-driven approaches have gained importance in sports broadcast, for example [2] learnt directorial styles by training classifiers on a training set of previous broadcasts. They suggest that such an approach could also be useful to determine the boundary of the salient events.…”

Section: Related Workmentioning

confidence: 99%

“…One approach to this problem is to isolate and process videos by genre. Such an approach has a twofold advantage: (1) each genre could be associated with a set of rules for video creation that might make it easier to design video understanding algorithms and (2) it is easier to distinguish between relevant and irrelevant semantic contents when information about genre is given (for example, information about crowds in a football match is rarely searched for and hence can be ignored). Recently, problems related to sports video analysis have particularly received much attention in this direction with many direct applications like automated highlights generation [1] or analysis of team activities and strategies [13], being built upon semantic analysis of video content.…”

Section: Introductionmentioning

confidence: 99%

Automatic analysis of broadcast football videos using contextual priors

Sharma

Gandhi

Chari

et al. 2016

SIViP

View full text Add to dashboard Cite

The presence of standard video editing practices in broadcast sports videos, like football, effectively means that such videos have stronger contextual priors than most generic videos. In this paper, we show that such information can be harnessed for automatic analysis of sports videos. Specifically, given an input video, we output per-frame information about camera angles and the events (goal, foul, etc.). Our main insight is that in the presence of temporal context (camera angles) for a video, the problem of event tagging (fouls, corners, goals, etc.) can be cast as per frame multiclass classification problem. We show that even with simple classifiers like linear SVM, we get significant improvement in the event tagging task when contextual information is included. We present extensive results for 10 matches from the recently concluded Football World Cup, to demonstrate the effectiveness of our approach.

show abstract

“…Chen et al [14] focused on features of a group of objects such as the number of players who are visible from a viewpoint. Daniyal et al presented an algorithm for viewpointquality ranking based on frame-level features, including size and location of the players in a basketball game [13], [25].…”

Section: Game Context Based Viewpoint Navigationmentioning

confidence: 99%

Personal Viewpoint Navigation Based on Object Trajectory Distribution for Multi-View Videos

Wang

Hara

Enokibori

et al. 2018

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

SUMMARYMulti-camera videos with abundant information and high flexibility are useful in a wide range of applications, such as surveillance systems, web lectures, news broadcasting, concerts and sports viewing. Viewers can enjoy an enhanced viewing experience by choosing their own viewpoint through viewing interfaces. However, some viewers may feel annoyed by the need for continual manual viewpoint selection, especially when the number of selectable viewpoints is relatively large. In order to solve this issue, we propose an automatic viewpoint navigation method designed especially for sports. This method focuses on a viewer's personal preference for viewpoint selection, instead of common and professional editing rules. We assume that different trajectory distributions of viewing objects cause a difference in the viewpoint selection according to personal preference. We learn the relationship between the viewer's personal viewpoint-selection tendency and the spatio-temporal game context represented by the objects trajectories. We compare three methods based on Gaussian mixture model, SVM with a general histogram and SVM with a bag-of-words to seek the best learning scheme for this relationship. The performance of the proposed methods are evaluated by assessing the degree of similarity between the selected viewpoints and the viewers' edited records. key words: multi-view video navigation, user preference, Gaussian mixture model

show abstract

Computational sports broadcasting: Automated director assistance for live sports

Cited by 25 publications

References 12 publications

Learning Sports Camera Selection From Internet Videos

Learning Sports Camera Selection From Internet Videos

Automatic analysis of broadcast football videos using contextual priors

Personal Viewpoint Navigation Based on Object Trajectory Distribution for Multi-View Videos

Contact Info

Product

Resources

About