Multiperspective Teaching of Unknown Objects via Shared-gaze-based Multimodal Human-Robot Interaction

Abstract

For successful deployment of robots in multifaceted situations, an understanding of the robot for its environment is indispensable. With advancing performance of state-of-the-art object detectors, the capability of robots to detect objects within their interaction domain is also enhancing. However, it binds the robot to a few trained classes and prevents it from adapting to unfamiliar surroundings beyond predefined scenarios. In such scenarios, humans could assist robots amidst the overwhelming number of interaction entities and impart the requisite expertise by acting as teachers. We propose a novel pipeline that effectively harnesses human gaze and augmented reality in a human-robot collaboration context to teach a robot novel objects in its surrounding environment. By intertwining gaze (to guide the robot's attention to an object of interest) with augmented reality (to convey the respective class information) we enable the robot to quickly acquire a significant amount of automatically labeled training data on its own. Training in a transfer learning fashion, we demonstrate the robot's capability to detect recently learned objects and evaluate the influence of different machine learning models and learning procedures as well as the amount of training data involved. Our multimodal approach proves to be an efficient and natural way to teach the robot novel objects based on a few instances and allows it to detect classes for which no training dataset is available. In addition, we make our dataset publicly available to the research community, which consists of RGB and depth data, intrinsic and extrinsic camera parameters, along with regions of interest.

Integrating Egocentric and Robotic Vision for Object Identification Using Siamese Networks and Superquadric Estimations in Partial Occlusion Scenarios

“…Continuing to explore alternative methods, [24] focused on a novel approach using human gaze and augmented reality (AR) in human-robot collaboration. Their method allows robots to identify and learn about unknown objects by acquiring automatically labeled training data, thereby enhancing their object detection capabilities.…”

Establishing Teacher-Student Communicative Empathy in a Multimodal Interactive Environment and Skill Enhancement

“…In the field of education, multi-modal interactive environment gives teachers and students more convenient ways of communication as their communication is no longer limited to the face-to-face communication, but can be conducted via multiple means including languages, body gestures, and facial expressions [5,6]. However, despite the increasing application of multi-modal interactive environment and its positive influence on the communicative empathy of teachers and students, it hasn't received enough attention from field scholars yet [7][8][9][10][11].…”