Keyframes, also known as I-Frames (Intra-Frames), are fundamental elements in video compression and encoding. They are complete, self-contained frames that do not rely on other frames for decoding.

This project aims to provide an educational exploration of keyframes and various types of frames in video processing, shedding light on their roles and operations within the multimedia landscape.

Abstract: Multimodal large language models (MLLMs) have enabled open-world visual understanding by injecting visual input as extra tokens into large language models (LLMs) as contexts. However, when ...