Using Nvidia Gaze SDK to Increase Video Engagement
The scarcity of face-to-face interaction in online engineering classes, compared to its face-to-face counterpart, makes it challenging to keep students engaged and motivated. One of the significant challenges in online learning is keeping students engaged while using pre-recorded videos. Pre-recorded videos often lack the personal touch of face-to-face interaction and do not allow for real-time interaction or feedback Al-Samarraie and Saeed (2019). In this paper, we explore the use of Nvidia Maxine as a tool to improve student engagement perception by processing pre-existing videos.
Nvidia Maxine is a software development kit (SDK) that uses eye-tracking technology to monitor the presenter’s eye movements and uses AI to process the video and redirect the presenter’s gaze toward the camera. This technology enables a more personalized experience for the viewer creating the feeling of a face-to-face conversation by creating a more immersive and engaging environment. The SDK provides APIs and libraries that enable course producers to automate this process and makes it scalable to process existing content in batches.
Conference recording will be available soon...
This is a work-in-progress paper in which we will discuss the process of automating video processing using Nvidia Maxine and the student’s perception by comparing the original videos to the processed ones. This work is innovative because it contributes to solving one of the significant challenges in online learning – keeping students engaged while using pre-recorded videos. This novel workflow is relevant to Engineering students as it has been shown in pay close attention to details Goldberg, J.R. and Barrett, A.F. (2005), and viewing videos where the professor’s gaze is looking elsewhere could contribute to distraction and cognitive load Lin and Cho (2019). Therefore we will consider this aspect as well when recording student perceptions.
Our approach of using Nvidia Maxine to redirect the presenter’s gaze towards the camera can create a more personalized experience for students, thus increasing engagement levels. This innovation is relevant to engineering education because it can improve the online learning experience for engineering students, who may have difficulty staying engaged with pre-existing videos.
Overall, our paper will demonstrate in the future how Nvidia Maxine can be used to process pre-existing videos and improve engagement levels in online education. Our work is relevant to engineering education and is innovative because it provides a solution to a significant challenge in online learning. We hope that this paper will encourage more research on the use of Nvidia Maxine in online education and inspire the development of new applications that leverage eye-tracking technology to enhance the online learning experience.
