As devices such as computers, sensors and motors have decreased in size, wearable computing has become an increasingly active area of research. Guidance of general human behavior by expert navigation is expected to be realized through the development of wearable computing. We aim to apply this technology to first-aid treatment. If an injured or ill person requires first-aid treatment but only non-experts are nearby, guidance from an expert at a remote site would be necessary. We focus on the sharing of visual information between a first-aid expert at a remote critical care center and a non-expert at the patient's location. Visual information about the remote site is captured by a head-mounted camera worn by the assistant and is shared with the expert, thus providing rich information on the status of the patient, the environment and the situation. In a typical master-slave configuration, the shared visual information is tightly linked between the expert and the assistant; that is, the assistant views what the expert wants to see. However, a tight link or synchronization is required only in certain situations, rather than at all times. The aim of this research is to develop a virtual viewing direction operation system with image stabilization for asynchronous visual information sharing between an expert and an assistant. The proposed system consists of three main parts: image stabilization, virtual viewing direction operation and the relative orientation display. The effectiveness of the developed system was confirmed experimentally.