Will 3D Visual Information Truly Become “Streamable” One Day?

Three-dimensional (3D) visual information holds broad application prospects in fields such as virtual reality and embodied intelligence, yet it has not achieved widespread adoption during its long-term technological evolution. In recent years, with the continuous advancement of light field video technology and the gradual integration of point cloud acquisition into everyday applications like autonomous driving—particularly as 3D Gaussian splatting demonstrates unprecedented performance advantages in rendering quality and speed—it naturally prompts the question: has the moment arrived for the widespread application of 3D visual information? This article will explore the key challenges faced by current technologies and standards in practical implementation from perspectives such as 3D visual information modeling and efficient encoding and transmission. It will also further reflect on the question: will 3D visual information truly become streamable one day.