This article proposes a generative adversarial network called explicit affine disentangled generative adversarial network (EAD-GAN), which explicitly disentangles affine transform in a self-supervised manner. We propose an affine transform regularizer to force the InfoGAN to have explicit properties of affine transform. To facilitate training an affine transform encoder, we decompose the affine matrix into two separate matrices and infer the explicit transform parameters by the least-squares method. Unlike the existing approaches, representations learned by the proposed EAD-GAN have clear physical meaning, where transforms, such as rotation, horizontal and vertical zooms, skews, and translations, are explicitly learned from training data. Thus, we set different values of each transform parameter individually to generate specifically affine transformed data by the learned network. We show that the proposed EAD-GAN successfully disentangles these attributes on the MNIST, CelebA, and dSprites datasets. EAD-GAN achieves higher disentanglement scores with a large margin compared to the state-of-the-art methods on the dSprites dataset. For example, on the dSprites dataset, EAD-GAN achieves the MIG and DCI score of 0.59 and 0.96 respectively, compared to 0.37 and 0.71, respectively, for the state-of-the-art methods.

Educational technological applications, such as computerized learning environments and robot tutors, are often programmed to provide social cues for the purposes of facilitating natural interaction and enhancing productive outcomes. However, there can be potential costs to social interactions that could run counter to such goals. Here, we present an experiment testing the impact of a watchful versus non-watchful robot tutor on children's language-learning effort and performance. Across two interaction sessions, children learned French and Latin rules from a robot tutor and filled in worksheets applying the rules to translate phrases. Results indicate better performance on the worksheets in the session in which the robot looked away from, as compared to the session it looked toward the child, as the child was filling in the worksheets. This was the case in particular for the more difficult worksheet items. These findings highlight the need for careful implementation of social robot behaviors to avoid counterproductive effects.

An algorithm for the robust detection and recognition of gestures for the interaction between human and a domestic floor cleaner robot is presented. The gestures are selected through a user study, in which the participants are asked to show natural gestures to the robot in given specific interaction scenarios. The gestures selected are those repeated by majority of participants and consist both commanding (say start cleaning) and social interaction (say greeting) gestures. The gesture recognition algorithm is developed using a combination of robust angular, positional, and directional features. The frontal and sagittal plains of human body are identified and invariant angular features are extracted from the skeletal data of a Kinect sensor. Robust positional and directional features are extracted by skeletal reconstruction using the invariant angular features and link rotation matrices. Dynamic time warping of features is done to make the algorithm robust to gesturing speed. Gestures are detected and recognized by calculating multiclass probability estimates by pairwise coupling method. The algorithm provided 97.26% recognition accuracy for a ten class robot commanding gesture database collected from multiple subjects.

A robust hand gesture detection and recognition algorithm using dynamic time warping and multi-class probability estimates is proposed. Quaternion based directional features of the hand are extracted using the color-depth camera Kinect. The directional features utilized have position and orientation invariance. Dynamic time warping of the signal sequence is done to achieve gesture size and speed invariance, and to enhance the gesture detection capability. The gestures are detected by hierarchical thresholding of the gesture probability and warping distance. Classification of gestures is done by multi-class probability estimates. The proposed algorithm is tested using a 12 class alphabet gesture database having variations in size, orientation, and speed. The algorithm provided 97.72% detection and 96.85% recognition accuracies respectively. A comparison of the proposed method with existing approaches (for detection as well as recognition) shows its better performance.

Computerized learning environments often include implementations of simulated social agents, incorporating the reasonable design assumption that learning interactions that are more social are more effective. However, recent evidence suggests that computerized social agents can in some circumstances fail to promote or even hinder learning. The current paper outlines evidence both supporting and arguing against the utility of computerized social agents for learning. We propose a framework reconciling this evidence by delineating impacts from a social agent that impinge upon different points of learning, from shallow to deep phases. Shallow social impacts when ineffectual carry little to no potential to actively impede learning, but any potential positive impacts on shallow learning phases are relatively limited in the absence of positive deep impacts. Deep social impacts, on the other hand, carry the potential to strongly drive deep learning, but when ineffectual carry the risk to impede it. The paper concludes with a proposal for future research based on this framework.

This paper provides a theoretical analysis of the feasibility of developing a robot that could take the role of an effective teacher by examining qualities and roles of effective teachers in the classrooms.

Teaching is inherently a social interaction between teacher and student. Despite this knowledge, many educational tools, such as vocabulary training programs, still model the interaction in a tutoring scenario as unidirectional knowledge transfer rather than a social dialog. Therefore, ongoing research aims to develop virtual agents as more appropriate media in education. Virtual agents can induce the perception of a life-like social interaction partner that communicates through natural modalities such as speech, gestures and emotional expressions. This effect can be additionally enhanced with a physical robotic embodiment.

Nonverbal behaviors serve as a rich source of information in inter human communication. In particular, motion cues can reveal details on a person's current physical and mental state. Research has shown, that people do not only interpret motion cues of humans in these terms, but also the motion of animals and inanimate devices such as robots. In order to successfully integrate mobile robots in domestic environments, designers have therefore to take into account how the device will be perceived by the user.

Autonomous robots tend to induce the perception of a personality through their behavior and appearance. It has been suggested that the personality of a robot can be used as a design guideline and as a mental model of the robot. We propose a method to design and evaluate personality and expressions for domestic robots.

Robots use their movement to interact with the user. Their movement is a crucial part in designing the interfaces of personal robots. It has a major impact on how a robot is perceived by the user interacting with it. Subtle flaws in the movement can create a repelling feeling and therefore affect negatively the interaction. On the other hand, carefully designed little quirks can give the robot personality and make it more interesting. A possible relationship between appearance and motion on how a device is perceived is described by the 'Uncanny Valley' conceived by Prof. M. Mori in 1970. It is widely used as a design guideline for creating humanoid interface robots. It does not, however, describe the impact that the quality of movements has on the perception and the believability of a character. Conclusive frameworks for the animation of robots are rare and are often adapted from cartoon or computer animation. This paper presents a refined design guideline for the motion of robotic characters with respect to the design requirements naturalness, adequateness and development over time. Furthermore, we created a set of tools that eases the process of designing the movement by improving on common animation techniques such as key-framed or scripted animations. The tools preserve the freedom of the animator to define specific expressive motions while at the same time offering the full power of scripting technology.