Michael Ryoo

11

Papers

120

Total Citations

Papers (11)

Go-with-the-Flow: Motion-Controllable Video Diffusion Models Using Real-Time Warped Noise

VicTR: Video-conditioned Text Representations for Activity Recognition

Mirasol3B: A Multimodal Autoregressive Model for Time-Aligned and Contextual Modalities

MAGICK: A Large-scale Captioned Dataset from Matting Generated Images using Chroma Keying

Adaptive Caching for Faster Video Generation with Diffusion Transformers

Learning to Localize Objects Improves Spatial Reasoning in Visual-LLMs

Learning Viewpoint-Agnostic Visual Representations by Recovering Tokens in 3D Space

Temporal Gaussian Mixture Layer for Videos

AViD Dataset: Anonymized Videos from Diverse Countries

TokenLearner: Adaptive Space-Time Tokenization for Videos

Does Self-supervised Learning Really Improve Reinforcement Learning from Pixels?