The field of digital telepresence and 3D facial animation is moving towards more realistic and accessible technologies. Researchers are exploring new methods to capture and replicate human movements and emotions, enabling more immersive and interactive experiences. One of the key areas of focus is the development of generative models that can reconstruct human appearance and movements from minimal input, such as egocentric views. Another important aspect is the creation of high-fidelity 3D head avatars that can accurately capture subtle facial expressions and emotions. Noteworthy papers in this area include EgoAnimate, which introduces a generative prior-based approach to reconstruct animatable avatars from egocentric inputs, and ScaffoldAvatar, which proposes a patch-based approach to create ultra-high fidelity, expressive, and photorealistic 3D head avatars. FantasyPortrait is also notable for its diffusion transformer-based framework that can generate high-fidelity and emotion-rich animations for both single- and multi-character scenarios.