Searching for Soccer Scenes using Siamese Neural Networks
Luca Reeb
In: Towards Data Science 2022
We have access to a large soccer database, containing a seasons worth of tracking-data, i.e. player trajectories, game statistics and expert-annotated events like pass or shot at goal from the German Bundesliga. While events allow you to find set-pieces like corner-kicks, the results are coarsely grained in that they do not consider how the players acted during the event. Also, some situations of potential interest, like counter attack, are not represented by an event. To enable fine-grained analysis of soccer matches, player movement (i.e. tracking-data) has to be considered.
Deep Siamese Metric Learning: A Highly Scalable Approach to Searching Unordered Sets of Trajectories
Löffler, C., Reeb, L., Dzibela, D., Marzilger, R., Witt, N., Eskofier, B. M., & Mutschler, C.
In: ACM Journal 2021
This work proposes metric learning for fast similarity-based scene retrieval of unstructured ensembles of trajectory data from large databases. We present a novel representation learning approach using Siamese Metric Learning that approximates a distance preserving low-dimensional representation and that learns to estimate reasonable solutions to the assignment problem. To this end, we employ a Temporal Convolutional Network architecture that we extend with a gating mechanism to enable learning from sparse data, leading to solutions to the assignment problem exhibiting varying degrees of sparsity.
Our experimental results on professional soccer tracking data provides insights on learned features and embeddings, as well as on generalization, sensitivity, and network architectural considerations. Our low approximation errors for learned representations and the interactive performance with retrieval times several magnitudes smaller shows that we outperform previous state of the art.
Effiziente Suche und Bewertung von Szenen in Spielsportarten
Reeb, L., Dzibela, D., Marzilger, R., & Witt, N.
In: spinfortec digital 2020
Durch tiefe neuronale Netze, sowie klassische Verfahren des maschinellen Lernens soll die Suche nach Szenen in Trackingdaten stark beschleunigt werden. Des Weiteren ist geplant, alternative/bessere Lösungen einer Spielsituation durch verstärkendes Lernen vorzuschlagen
IALE: Imitating Active Learner Ensembles.
Löffler, C., & Mutschler, C. (2022).
In: Vortrag NeurIPS New Orleans (2022)
Sports Scene Searching, Rating & Solving using AI.
Marzilger, R., Hirn, F., Alvarez R.A. & Witt, N. (2022).
https://nbn-resolving.org/urn:nbn:de:bsz:ch1-qucosa2-807718
In: Krumm, D., Schwanitz, S., & Odenwald, S. (2022). spinfortec2022 : Tagungsband zum 14. Symposium der Sektion Sportinformatik und Sporttechnologie der Deutschen Vereinigung für Sportwissenschaft (dvs) Chemnitz 29. - 30. September 2022.
IALE: Imitating Active Learner Ensembles.
Löffler, C., & Mutschler, C. (2022).
http://jmlr.org/papers/v23/21-0387.html
In: Journal of Machine Learning Research, 23(107), 1–29.
Don’t Get Me Wrong: How to apply Deep Visual Interpretations to Time Series.
Loeffler, C., Lai, W.-C., Eskofier, B., Zanca, D., Schmidt, L., & Mutschler, C. (2022).
https://doi.org/10.48550/ARXIV.2203.07861
In: arxiv.org
Active Learning of Ordinal Embeddings: A User Study on Football Data.
Loeffler, C., Fallah, K., Fenu, S., Zanca, D., Eskofier, B., Rozell, C. J., & Mutschler, C. (2022).
https://doi.org/10.48550/ARXIV.2207.12710
In: arxiv.org