In this work, we present a method for automatically generating a synthetic, labelled computer vision dataset of game-character positions for LoL and the application of this synthetic data for training a deep-learning, single-shot object detection model that can then be applied in pseudo labelling and generating detailed player position data from LoL video streams. Furthermore, we investigate the effects of data augmentation and class balancing on the overall performance of the neural network, comparing them to the base network trained on a manually labelled dataset. Using synthetically generated datasets of 10000 images containing 21 object classes, we achieved a mean average precision (mAP@50) of 64.8% and an IoU of 67.7%.
In this work, we present a method for automatically generating a synthetic, labelled computer vision dataset of game-character positions for LoL and the application of this synthetic data for training a deep-learning, single-shot object detection model that can then be applied in pseudo labelling and generating detailed player position data from LoL video streams. Further-more, we investigate the effects of data augmentation and class balancing on the overall performance of the neural network, comparing them to the base network trained on a manually labelled dataset. Using synthetically generated datasets of 10000 images containing 21 object classes, we achieved a mean average precision (mAP@50) of 64.8% and an IoU of 67.7%.
Data Pipelines for Real-Time, Custom Object Detection and Tracking in League of Legends / Mutsvanga, D.; Petrosyants, A.; Nikolaev, D.; Orlova, J.; Stepanov, A.; Passerone, R.; Somov, A.. - (2024), pp. 1-6. ( 19th IEEE Sensors Applications Symposium, SAS 2024 Italy 2024) [10.1109/SAS60918.2024.10636474].
Data Pipelines for Real-Time, Custom Object Detection and Tracking in League of Legends
Stepanov A.;Passerone R.;Somov A.
2024-01-01
Abstract
In this work, we present a method for automatically generating a synthetic, labelled computer vision dataset of game-character positions for LoL and the application of this synthetic data for training a deep-learning, single-shot object detection model that can then be applied in pseudo labelling and generating detailed player position data from LoL video streams. Furthermore, we investigate the effects of data augmentation and class balancing on the overall performance of the neural network, comparing them to the base network trained on a manually labelled dataset. Using synthetically generated datasets of 10000 images containing 21 object classes, we achieved a mean average precision (mAP@50) of 64.8% and an IoU of 67.7%.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione
