Flow is an affective state of optimal experience, total immersion and high productivity. While often associated with (professional) sports, it is a valuable information in several scenarios ranging from work environments to user experience evaluations, and we expect it to be a potential reward signal for human-in-the-loop reinforcement learning systems. Traditionally, flow has been assessed through questionnaires which prevents its use in online, real-time environments. In this work, we present our findings towards estimating a user's flow state based on physiological signals measured using wearable devices. We conducted a study with participants playing the game Tetris in varying difficulty levels, leading to boredom, stress, and flow. Using an end-to-end deep learning architecture, we achieve an accuracy of 67.50% in recognizing high flow vs. low flow states and 49.23% in distinguishing all three affective states boredom, flow, and stress.

In recent years, the importance of location-based services and indoor positioning systems increased significantly for both, research and industry. Visual localization systems have the advantage of not depending on dedicated infrastructure and thus they are interesting for navigation within buildings. While there are already approaches which are using pre-recorded databases of reference images to obtain an absolute position for a given query image, suitable applications which are estimating the relative movement of pedestrians out of a first person perspective video are still missing. This paper presents a novel approach for a pedometer as well as for an activity detector using a such a first person perspective video stream of a pedestrian as input data. The system counts the number of steps and furthermore detects current activities of a user. Therefore, we analyze all video input data with the SURF algorithm in order to extract robust feature points. Especially the orientation and scaling properties of this feature points are used for an accurate measurement.

In recent years, location-based services and indoor positioning systems gained increasing importance for both, research and industry. Visual localization systems have the advantage of not being dependent on dedicated infrastructure and thus are especially interesting for navigation within buildings. While there are already approaches of using pre-recorded databases of reference images to obtain an absolute position for a given query image, suitable means to estimate the relative movement of pedestrians from an ego perspective video are still missing. This paper presents a novel visual odometry system for pedestrians. The user carries a mobile device while walking - the camera aims into the direction of walking. Using only the video stream as input, the system generates a two-dimensional trajectory, which describes the path traveled by the user. Both, the user's current heading as well as the walking direction are estimated based on the movement of visual feature points in successive video frames. In order to assess the accuracy of the system, it is evaluated in three different scenarios (indoors in an university building, in an urban area and in a city park). Not relying on reference points (for instance provided by a database, which references visual feature points with geo-data), the error accumulates with distance traveled. After a walked distance of 100 meters, the average error lies between 4.6 and 13.9 meters (depending on the scenario). Consequently, the system is a promising approach for visual odometry, which can be used in conjunction with existing absolute visual positioning systems or as a core part of a future SLAM (simultaneous localization and mapping) system.

Due to the increasing popularity of location-based services, the need for reliable and cost-effective indoor positioning methods is rising. As an alternative to radio-based localization methods, in 2011, we introduced MoVIPS (Mobile Visual Indoor Positioning System), which is based on the idea to extract visual feature points from a query image and compare them to those of previously collected geo-referenced images. The general feasibility of positioning by SURF points on a conventional smartphone was already shown in our previous work. However, the system still faced several shortcomings concerning real-world usage such as request times being too high and distance estimation being unreliable because of the employed estimation method not being rotation invariant. In this paper, three extensions are presented that improve the practical applicability of MoVIPS. To speed up request times, both a dead reckoning approach (based on step counting using the accelerometer) and an orientation estimation (based on the smartphones compass) are introduced to filter relevant images from the database and thus to reduce the amount of images to compare the query image to. Furthermore, the vectors of the SURF points are quantized. For this purpose, clusters are calculated from all SURF points from the database. As a result, each image can be represented by a histogram of cluster frequencies, which can be compared with each other a lot more efficiently. The third extension is an improvement of the distance estimation method, which uses the matched feature points of an image to perform a perspective transformation and to determine the actual position with the aid of the transformation matrix.

In this paper, we investigate pointing as a lightweight form of gestural interaction in cars. In a pre-study, we show the technical feasibility of reliable pointing detection with a depth camera by achieving a recognition rate of 96% in the lab. In a subsequent in-situ study, we let drivers point to objects inside and outside of the car while driving through a city. In three usage scenarios, we studied how this influenced their driving objectively, as well as subjectively. Distraction from the driving task was compensated by a regulation of driving speed and did not have a negative influence on driving behaviour. Our participants considered pointing a desirable interaction technique in comparison to current controller-based interaction and identified a number of additional promising use cases for pointing in the car.

With the increasing computational power of mobile devices and the increase in the usage of ordinary location-based services, the area of indoor location-based services is of growing interest. Nowadays indoor location-based services are used mainly for personalized information retrieval of maps and points of interest. Advanced location-based functionality often suffers from imprecise positioning methods. In this paper we present a simple, yet powerful positioning method inside buildings which allows for a fine-grained detection of the position and orientation of a user while being easy to deploy and optimize. The main contribution of this paper consists of the combination of an image recognition system with a distance estimation algorithm to gain a high-quality positioning service independent from any infrastructure using the camera of a mobile device. Moreover this type of positioning can be operated in a user-contributed way and is less susceptible to small changes in the environment as compared to popular WLAN-based systems. As an extension, we propose the usage of very coarse WLAN positioning to reduce the size of the candidate set of image recognition and hence speed up the system.