Abstract We present initial results of the study where we used microphones, placed in front of nest boxes, to monitor daily foraging activity of bumblebees. Sound recordings were analyzed using a custom-made computer algorithm which detects flight buzzing sounds coming from arrivals or departures of individual bees. In addition, the algorithm distinguishes between arrivals and departures. We show examples of daily activities for three... We present initial results of the study where we used microphones, placed in front of nest boxes, to monitor daily foraging activity of bumblebees. Sound recordings were analyzed using a custom-made computer algorithm which detects flight buzzing sounds coming from arrivals or departures of individual bees. In addition, the algorithm distinguishes between arrivals and departures. We show examples of daily activities for three species (B. pascuorum, B. humilis and B. hypnorum), each was monitored over the course of one day. This paper presents initial results of a longer study where we plan to systematically investigate the activities of bumblebees in various circumstances. Read More Read Less

Abstract Physiological signals give important insight regarding some one’s health. It would be in the interest of people to monitor such signals without any wearable devices. We used RGB camera recordings of faces to reconstruct the PPG signal, which can be used to monitor many physiological signals such as heart rate, breathing rate, blood pressure, etc. A deep learning method was developed... Physiological signals give important insight regarding some one’s health. It would be in the interest of people to monitor such signals without any wearable devices. We used RGB camera recordings of faces to reconstruct the PPG signal, which can be used to monitor many physiological signals such as heart rate, breathing rate, blood pressure, etc. A deep learning method was developed to enhance existing state-of-the-art methods. This method uses the output of an existing method as an input into a LSTM neural network, which substantially improves the reconstruction of PPG. Read More Read Less

Abstract Emotion recognition is an important part of affective-aware applications. Specifically, using audio speech signal has the advantage of being compatible with applications using a natural language interface. There are multiple valid representations of emotions. We propose a new representation aimed at using differently labeled databases jointly. We include a short overview of some of the available databases and methods for feature extraction... Emotion recognition is an important part of affective-aware applications. Specifically, using audio speech signal has the advantage of being compatible with applications using a natural language interface. There are multiple valid representations of emotions. We propose a new representation aimed at using differently labeled databases jointly. We include a short overview of some of the available databases and methods for feature extraction and selection. Both classification of emotions and regression in 2D emotional space are discussed. We concentrate on using neural networks for both tasks. Regression provides good results but is hard to interpret while classification is more robust. Read More Read Less