Immersive Video Experience: 360VR video and beyond that, 6DOF video will provide users with an immersive media experience in the future. People will want a more realistic and three-dimensional environment and will consume more realistic media. If so, how much compression is needed to transmit these much larger and heavier videos to smartphones and VR devices? How sensitive is the human eye to the image quality of immersive media?

Efficient Deep Learning Model: Now, deep learning research is also expanding the content range from simple pictures to videos, and deep learning models are getting deeper and wider. While this contributes to increasing the accuracy of image classification/interpretation/prediction, it is difficult to use in real life because it requires an expensive and huge device. Therefore, less complex deep learning models will be needed for power-limited devices such as smartphones/cars/drones/IoT, which are easily accessible by people. Accordingly, we explore whether deep learning models are also compressible and can be modeled faster.