Tianyu He

14

Papers

128

Total Citations

Papers (14)

GAIA: Zero-shot Talking Avatar Generation

3D-LLaVA: Towards Generalist 3D LMMs with Omni Superpoint Transformer

InstructAvatar: Text-Guided Emotion and Motion Control for Avatar Generation

To Grok or not to Grok: Disentangling Generalization and Memorization on Corrupted Algorithmic Datasets

(How) Can Transformers Predict Pseudo-Random Numbers?

HiFace: High-Fidelity 3D Face Reconstruction by Learning Static and Dynamic Details

Image Coding for Machines with Omnipotent Feature Learning

VidTwin: Video VAE with Decoupled Structure and Dynamics

Is Vanilla MLP in Neural Radiance Field Enough for Few-shot View Synthesis?

Partial Person Re-Identification With Part-Part Correspondence Learning

Cloth-Changing Person Re-Identification From a Single Image With Gait Prediction and Regularization

Dense Interaction Learning for Video-Based Person Re-Identification

Layer-Wise Coordination between Encoder and Decoder for Neural Machine Translation

Critical Initialization of Wide and Deep Neural Networks using Partial Jacobians: General Theory and Applications