Global terrorist threats and illegal migration have intensified concerns for the security of citizens, and every effort is made to exploit all available technological advances to prevent adverse events and protect people and their property. Due to the ability to use at night and in weather conditions where RGB cameras do not perform well, thermal cameras have become an important component of sophisticated video surveillance systems. In this paper, we investigate the task of automatic person detection in thermal images using convolutional neural network models originally intended for detection in RGB images. We compare the performance of the standard state-of-the-art object detectors such as Faster R-CNN, SSD, Cascade R-CNN, and YOLOv3, that were retrained on a dataset of thermal images extracted from videos that simulate illegal movements around the border and in protected areas. Videos are recorded at night in clear weather, rain, and in the fog, at different ranges, and with different movement types. YOLOv3 was significantly faster than other detectors while achieving performance comparable with the best, so it was used in further experiments. We experimented with different training dataset settings in order to determine the minimum number of images needed to achieve good detection results on test datasets. We achieved excellent detection results with respect to average accuracy for all test scenarios although a modest set of thermal images was used for training. We test our trained model on different well known and widely used thermal imaging datasets as well. In addition, we present the results of the recognition of humans and animals in thermal images, which is particularly important in the case of sneaking around objects and illegal border crossings. Also, we present our original thermal dataset used for experimentation that contains surveillance videos recorded at different weather and shooting conditions.

Many computer vision applications rely on accurate and fast object detection, and in our case, ball detection serves as a prerequisite for action recognition in handball scenes. We compare the performance of two of the state-of-the-art convolutional neural network-based object detectors for the task of ball detection in non-staged, real-world conditions. The comparison is performed in terms of speed and accuracy measures on a dataset comprising custom handball footage and a sample of images obtained from the Internet. The performance of the models is compared with and without additional training with examples from our dataset.

This paper considers the problem of tracking the players in handball videos using a single video source. Tracking of handball players in the video is a difficult task as they can frequently leave and re-enter the camera field of view, often change directions quickly and occlude each other. Players wear similar team uniforms, play indoor under artificial illumination, with the background than can vary significantly as the handball court is often painted in various colors. The continually improving accuracy of CNN-based object detectors makes tracking-by-detection methods an attractive choice for tracking players in sports videos as they can perform online and with low computational requirements on top of object detection. Here we consider the use of three tracking-by-detection methods in conjunction with the YOLO object detector, namely the standard Hungarian assignment algorithm, the Simple Online, and Real-time Tracking (SORT) algorithm that adds a motion model, and its extension Deep SORT. The methods are tested on a custom dataset of handball video scenes.

In team sports training scenes, it is common to have many players on the court, each with his own ball performing different actions. Our goal is to detect all players in the handball court and determine the most active player who performs the given handball technique. This is a very challenging task, for which, apart from an accurate object detector, which is able to deal with complex cluttered scenes, additional information is needed to determine the active player. We propose an active player detection method that combines the Yolo object detector, activity measures, and tracking methods to detect and track active players in time. Different ways of computing player activity were considered and three activity measures are proposed based on optical flow, spatiotemporal interest points, and convolutional neural networks. For tracking, we consider the use of the Hungarian assignment algorithm and the more complex Deep SORT tracker that uses additional visual appearance features to assist the assignment process. We have proposed the evaluation measure to evaluate the performance of the proposed active player detection method. The method is successfully tested on a custom handball video dataset that was acquired in the wild and on basketball video sequences. The results are commented on and some of the typical cases and issues are shown.