Next-Generation Video Coding

Oct 1, 2024

With the continuous development of generative models (diffusion), this project explores a new video encoding paradigm starting from ultra-low bit-rates, gradually resolving bottlenecks and promoting next-generation video coding technology.

Related work includes Bi-AGMI, a bidirectional frame interpolation model using diffusion models for video compression and transmission. (IEEE GCCE 2025 Oral)

Also includes FRS, a video slicing algorithm driven by motion intensity analysis. (IEVC 2026)

This research is conducted in collaboration with NICT Japan (情報通信研究機構).

Deep Learning Video Generation
ZIYUE ZENG
ZIYUE ZENG
Master’s Student → Incoming PhD Student at Waseda University

Pursuing socially meaningful, grounded research with enduring passion and diligence.