Hongyu Liu
  • About
  • Publications
  • Projects (current)
  • Memories

Selected Work

Projects

A few industry and research projects around digital humans, avatar animation, and controllable content creation.

Research & Industry

Project

DyaPlex

Full-duplex speech-motion model for dyadic interaction, enabling a digital agent to speak, listen, perceive partner motion, and generate responsive motion in a streaming loop.

Open project

Project

LongCat-Video-Avatar 1.5

Open-source audio-driven avatar video generation with stronger stability, lip-sync accuracy, identity consistency, and fast 8-step inference.

Open project

Project

OPSD-V

On-policy self-distillation for post-training few-step autoregressive video generators, improving long-horizon quality and motion dynamics while preserving the original sampler.

Open project
project thumbnail

Project

Anime Talking Heads

Anime talking heads. Control anime avatar with your expression in real time. We also build SDK, which only occupy 9% I7 CPU's performance (No GPU!).

View details
project thumbnail

Project

Consumer Digital Human.

Given a single image(T pose), we can create an avatar and drive this avatar with target motion.

Open project
project thumbnail

Project

Paperman

We can drive a 2D image with skeleton.

View details
© Copyright 2026 Hongyu Liu. Powered by Jekyll with al-folio theme. Hosted by GitHub Pages. Photos from Unsplash.