The goal of research in quantifying the visual quality of experience (QoE) is to develop objective measures that can automatically predict perceived visual quality, immersion, reality, presence, etc. Such objective metrics can play an important role in a broad range of applications, such as content production, acquisition, compression, communication, displaying, printing, restoration, enhancement, analysis, and watermarking.

Image & video restoration, enhancement, and manipulation are key computer vision tasks, aiming at the restoration of degraded image content, the filling in of missing information, or the needed transformation and/or manipulation to achieve the desired target (with respect to perceptual quality, contents, or performance of apps working on such images). Recent years have witnessed an increased interest from the vision and graphics communities in these fundamental topics of research.  

Deep generative models are the learned high-dimensional probability distributions via a huge number of observable samples by utilizing neural networks, which can be used to estimate the likelihood of each observation and to create new samples. Developing deep generative models has become one of the hottest researched fields in artificial intelligence in recent years. We are studying the recent approaches to construct deep generative models (e.g., VAE, VQ-VAE, NF, diffusion models, GANs) and applying them to several applications.

Visual object tracking is an important research topic in computer vision, image understanding, and pattern recognition. Given the initial state (center location and scale) of a target in the first frame of a video sequence, the aim of visual object tracking is to automatically obtain the states of the object in the subsequent video frames. Tracking the partially or even fully occluded object is challenging, thus now we are investigating the technique to achieve promising performance when even a target object is invisible.

The metaverse industry is rapidly emerging, and the users expect an increasingly high quality of experience for immersive content. Although the expanded virtual environment can deliver an improved sensation of reality to users, its quantitative positive/negative effect was rarely studied since the difficulties of human visual perception. We investigate which realistic content factors dominate human perceptual quality increases.