We present an mbt moja important aspect of our human-robot communication interface which is being developedin the context of our long-term research framework PERSES dealing with highly interactive mobile companionrobots. Based on a multi-modal mbt shop uk people detection and tracking system, we present a hierarchical neural architec-ture that estimates a target point at the mbt shop london floor indicated mbt shoes john lewis uk by a pointing pose, thus enabling a user to navigate a mo-bile robot to a specific target position in his local surroundings by means of pointing.
In this context, we wereespecially interested in determining whether it is possible to accomplish such a target point estimator using onlymonocular images of low-cost cameras. The estimator mbt boots canada has been implemented and experimentally investigated onour mobile robotic assistant… HOROS. Although only monocular image data of relatively poor quality were util-ized, the estimator accomplishes a good estimation mbt outlet singapore performance, achieving an accuracy better than that of a hu-man viewer on the same data.
The achieved recognition results demonstrate that it is in fact possible to realize auser-independent pointing direction estimation using monocular images only, but further efforts are necessary toimprove the robustness of this approach for everyday application.