IMPROVED VIDEO SUPER RESOLUTION

Summary from the TLDR AI newsletter

Researchers have developed a training-free video interpolation method for generative video diffusion models. This new approach, compatible with various models, enhances frame rates without the need for extensive training or large datasets.