a approach of query audience’s attention in virtual speech

Abstract Eye contact between speakers and audiences can help speakers get better effectiveness. virtual speeches are a very popular way for multi-user to communicate remotely, but it has the disadvantage of being unable to make eye contact. This paper proposes the online audience’s attention evaluation based on gaze tracking. Instead of using less-obtainable and expensive eye-tracking equipment, this article only uses webcams to capture audience's head pose, fixation time and other features. We also proposed a set of indexes, which can be used to evaluate the audience's degree of attention, making up for the defect that the speaker cannot obtain the audience's concentration through eye contact during online speeches. We selected 96 students and divided them into three groups for a 20-minute curriculum simulation, and used the Spearman correlation coefficient to analyse the correlation between our evaluation indicators and concentration, the result showed that each evaluation index has a significant correlation with the degree of attention (p=0.01), and all the students in the focused group met the thresholds set by each of our evaluation indicators, while the students in the non-focused group failed to reach the standard, the accuracy rate is 100%. In addition, we used EEG signals to verify the accuracy of the system evaluation.During the simulation, eye movement data and EEG signals were measured synchronously for the second group of students. The EEG results of the second group of students were consistent with the results of systematic evaluation.All the experimental work on the ordinary equipment such as a notebook computer which integrated a cpu of 2.30GHZ and a 130W pixel camera without any additional equipment. Our research provides a low-cost method for attention monitoring with reference value across multiple domains. In the field of webpage design, it can enhance page structure and improve user experience based on the distribution of user attention. In the realm of driving safety, real-time monitoring of driver attention can prevent fatigue driving. In the medical field, monitoring the attention of doctors and patients contributes to improving surgical safety. In education, attention monitoring of students allows for adjustments in teaching methods, ultimately enhancing the quality of education. In the industrial sector, monitoring worker attention can reduce operational errors and improve overall work efficiency.