Video Call

Video call technology is undergoing rapid advancements driven by the need for improved quality, security, and understanding of user behavior within virtual interactions. Current research focuses on enhancing video quality in real-time using efficient neural networks like recurrent bottleneck mixers, accurately predicting perceived video quality by incorporating temporal distortions, and detecting sophisticated deepfake impersonations through novel active probing techniques leveraging corneal reflections. These improvements are crucial for enhancing the reliability and security of video conferencing, impacting both user experience and the broader field of computer vision and security.