Christian Thurau

Pose primitive based human action recognition in videos or still images

This paper presents a method for recognizing human actions based on pose primitives. In learning mode, the parameters representing poses and activities are estimated from videos. In run mode, the method can be used both for videos or still images. For recognizing pose primitives, we extend a Histogram of Oriented Gradient (HOG) based descriptor to better cope with articulated poses and cluttered background. Action classes are represented by histograms of poses primitives. For sequences, we incorporate the local temporal context by means of n-gram expressions. Action recognition is based on a simple histogram comparison. Unlike the mainstream video surveillance approaches, the proposed method does not rely on background subtraction or dynamic features and thus allows for action recognition in still images.

Guns, swords and data: Clustering of player behavior in computer games in the wild

Behavioral data from computer games can be exceptionally high-dimensional, of massive scale and cover a temporal segment reaching years of real-time and a varying population of users. Clustering of user behavior provides a way to discover behavioral patterns that are actionable for game developers. Interpretability and reliability of clustering results is vital, as decisions based on them affect game design and thus ultimately revenue. Here case studies are presented focusing on clustering analysis applied to high-dimensionality player behavior telemetry, covering a combined total of 260,000 characters from two major commercial game titles: the Massively Multiplayer Online Role-Playing Game Tera and the multi-player strategy war game Battlefield 2: Bad Company 2. K-means and Simplex Volume Maximization clustering were applied to the two datasets, combined with considerations of the design of the games, resulting in actionable behavioral profiles. Depending on the algorithm different insights into the underlying behavior of the population of the two games are provided.

Early drought stress detection in cereals: simplex volume maximisation for hyperspectral image analysis

Early water stress recognition is of great relevance in precision plant breeding and production. Hyperspectral imaging sensors can be a valuable tool for early stress detection with high spatio-temporal resolution. They gather large, high dimensional data cubes posing a significant challenge to data analysis. Classical supervised learning algorithms often fail in applied plant sciences due to their need of labelled datasets, which are difficult to obtain. Therefore, new approaches for unsupervised learning of relevant patterns are needed. We apply for the first time a recent matrix factorisation technique, simplex volume maximisation (SiVM), to hyperspectral data. It is an unsupervised classification approach, optimised for fast computation of massive datasets. It allows calculation of how similar each spectrum is to observed typical spectra. This provides the means to express how likely it is that one plant is suffering from stress. The method was tested for drought stress, applied to potted barley plants in a controlled rain-out shelter experiment and to agricultural corn plots subjected to a two factorial field setup altering water and nutrient availability. Both experiments were conducted on the canopy level. SiVM was significantly better than using a combination of established vegetation indices. In the corn plots, SiVM clearly separated the different treatments, even though the effects on leaf and canopy traits were subtle.

How players lose interest in playing a game: An empirical study based on distributions of total playing times

Analyzing telemetry data of player behavior in computer games is a topic of increasing interest for industry and research, alike. When applied to game telemetry data, pattern recognition and statistical analysis provide valuable business intelligence tools for game development. An important problem in this area is to characterize how player engagement in a game evolves over time. Reliable models are of pivotal interest since they allow for assessing the long-term success of game products and can provide estimates of how long players may be expected to keep actively playing a game. In this paper, we introduce methods from random process theory into game data mining in order to draw inferences about player engagement. Given large samples (over 250,000 players) of behavioral telemetry data from five different action-adventure and shooter games, we extract information as to how long individual players have played these games and apply techniques from lifetime analysis to identify common patterns. In all five cases, we find that the Weibull distribution gives a good account of the statistics of total playing times. This implies that an average players interest in playing one of the games considered evolves according to a non-homogeneous Poisson process. Therefore, given data on the initial playtime behavior of the players of a game, it becomes possible to predict when they stop playing.

Learning Human-Like Opponent Behavior for Interactive Computer Games

Compared to their ancestors in the early 1970s, present day computer games are of incredible complexity and show magnificent graphical performance. However, in programming intelligent opponents, the game industry still applies techniques developed some 30 years ago. In this paper, we investigate whether opponent programming can be treated as a problem of behavior learning. To this end, we assume the behavior of game characters to be a function that maps the current game state onto a reaction. We will show that neural networks architectures are well suited to leam such functions and by means of a popular commercial game we demonstrate that agent behaviors can be learned from observation.

Analyzing the Evolution of Social Groups in World of Warcraft

This paper investigates the evolution of social structures in the game WORLD OF WARCRAFT®. We analyze 192 million recordings of 18 million characters belonging to 1.4 million teams, spanning a period of 4 years. Using a recent matrix factorization method, we extract lower dimensional data embeddings. The embeddings provide intuitively interpretable categorizations and we find a tendency towards guilds comprised of casual gamers. To our knowledge, this is the first study considering such a vast amount of data for analyzing groups in MMORPGs.

Descriptive matrix factorization for sustainability Adopting the principle of opposites

Climate change, the global energy footprint, and strategies for sustainable development have become topics of considerable political and public interest. The public debate is informed by an exponentially growing amount of data and there are diverse partisan interest when it comes to interpretation. We therefore believe that data analysis methods are called for that provide results which are intuitively understandable even to non-experts. Moreover, such methods should be efficient so that non-experts users can perform their own analysis at low expense in order to understand the effects of different parameters and influential factors. In this paper, we discuss a new technique for factorizing data matrices that meets both these requirements. The basic idea is to represent a set of data by means of convex combinations of extreme data points. This often accommodates human cognition. In contrast to established factorization methods, the approach presented in this paper can also determine over-complete bases. At the same time, convex combinations allow for highly efficient matrix factorization. Based on techniques adopted from the field of distance geometry, we derive a linear time algorithm to determine suitable basis vectors for factorization. By means of the example of several environmental and developmental data sets we discuss the performance and characteristics of the proposed approach and validate that significant efficiency gains are obtainable without performance decreases compared to existing convexity constrained approaches.

Face Detection Using GPU-Based Convolutional Neural Networks

In this paper, we consider the problem of face detection under pose variations. Unlike other contributions, a focus of this work resides within efficient implementation utilizing the computational powers of modern graphics cards. The proposed system consists of a parallelized implementation of convolutional neural networks (CNNs) with a special emphasize on also parallelizing the detection process. Experimental validation in a smart conference room with 4 active ceiling-mounted cameras shows a dramatic speed-gain under real-life conditions.

Predicting player churn in the wild

Free-to-Play or “freemium” games represent a fundamental shift in the business models of the game industry, facilitated by the increasing use of online distribution platforms and the introduction of increasingly powerful mobile platforms. The ability of a game development company to analyze and derive insights from behavioral telemetry is crucial to the success of these games which rely on in-game purchases and in-game advertising to generate revenue, and for the company to remain competitive in a global marketplace. The ability to model, understand and predict future player behavior has a crucial value, allowing developers to obtain data-driven insights to inform design, development and marketing strategies. One of the key challenges is modeling and predicting player churn. This paper presents the first cross-game study of churn prediction in Free-to-Play games. Churn in games is discussed and thoroughly defined as a formal problem, aligning with industry standards. Furthermore, a range of features which are generic to games are defined and evaluated for their usefulness in predicting player churn, e.g. playtime, session length and session intervals. Using these behavioral features, combined with the individual retention model for each game in the dataset used, we develop a broadly applicable churn prediction model, which does not rely on game-design specific features. The presented classifiers are applied on a dataset covering five free-to-play games resulting in high accuracy churn prediction.

Game Data Mining

During the years of the Information Age, technological advances in the computers, satellites, data transfer, optics, and digital storage has led to the collection of an immense mass of data on everything from business to astronomy, counting on the power of digital computing to sort through the amalgam of information and generate meaning from the data. Initially, in the 1970s and 1980s of the previous century, data were stored on disparate structures and very rapidly became overwhelming. The initial chaos led to the creation of structured databases and database management systems to assist with the management of large corpuses of data, and notably, the effective and efficient retrieval of information from databases. The rise of the database management system increased the already rapid pace of information gathering.