This is a collection of our NAS and Vision Transformer work.

Meta-Transformer for Unified Multimodal Learning

Open-source deep-learning framework for building, training, and fine-tuning deep learning models using state-of-the-art Physics-ML methods

Initial public release of code, data, and model weights for FourCastNet

Artificial Intelligence Research for Science (AIRS)

A benchmark for the next generation of data-driven global weather models.

A library for Koopman Neural Operator with Pytorch.

DeepONet & FNO (with practical extensions)

This is the python code corresponding to the article "Let You See in Haze and Sandstorm: Two-in-One Low-visibility Enhancement Network".

[TITS 2023] Dataset-Asynchronous Trajectory Matching-based Multimodal Maritime Data Fusion for Vessel Traffic Surveillance in Inland Waterways

[CVPRW 2023] SCANet: Self-Paced Semi-Curricular Attention Network for Non-Homogeneous Image Dehazing

[ECCV 2024] OneRestore: A Universal Restoration Framework for Composite Degradation

Double Domains-Guided Real-time Low-light Image Enhancement for UHD Transportation Surveillance

This is the python code corresponding to the article "Deep learning-driven surveillance quality enhancement for maritime management promotion under low-visibility weathers ".