Jialun Liu

Achieving state-of-the-art performance in both 4D reconstruction and point tracking, Uni4R leverages continuous velocity fields to model dynamics at any timestamp, breaking free from traditional limitations.

Ditch the multi-step optimization and spatial alignment headaches: GarmentPainter synthesizes stunningly realistic 3D garment textures with a simple, efficient character-guided diffusion approach.

By completing sparse kinematic data with a video diffusion Transformer, LiftAvatar unlocks more expressive and robust 3D avatar animation from monocular video, even for extreme and unseen expressions.

Achieve superior scene-consistent video generation by training a model to iteratively estimate geometry and render novel views, using geometry as context.