James Matthew Rehg

Full Professor@University of Illinois at Urbana-Champaign·美国·OpenReview

研究方向

Machine Learning · Deep Learning · Probabilistic Graphical Models · Continual Learning · Few Shot Learning · Computer Vision · First Person Vision · Action Recognition · Attention Estimation · Shape Estimation

James Matthew Rehg

DiffVax: Optimization-Free Image Immunization Against Diffusion-Based Editing

Self-Supervised Dynamical System Representations for Physiological Time-Series

Leveraging Shared Prototypes for a Multimodal Pulse Motion Foundation Model

Self-supervised Learning for Incomplete Multimodal Wearable Sensor Data

Cue3D: Quantifying the Role of Image Cues in Single-Image 3D Generation

Toward Human Deictic Gesture Target Estimation

DiffEye: Diffusion-Based Continuous Eye-Tracking Data Generation Conditioned on Natural Images

Fine-Grained Preference Optimization Improves Spatial Reasoning in VLMs

RelCon: Relative Contrastive Learning for a Motion Foundation Model for Wearable Data

DiffVax: Optimization-Free Image Immunization Against Diffusion-Based Editing

What is the Visual Cognition Gap between Humans and Multimodal LLMs?

Benchmarking Visual Cognition of Multimodal LLMs via Matrix Reasoning

AnyECG: Foundational Models for Electrocardiogram Analysis