Zhu Lab

DOFA

Code for Neural Plasticity-Inspired Foundation Model for Observing the Earth Crossing Modalities

softcon

Multi-label guided soft contrastive learning for efficient Earth observation pretraining

FloodCastBench

ST-FSOD

M3R-CR

High-Resolution Cloud Removal with Multi-Modal and Multi-Resolution Data Fusion: A New Baseline and Benchmark

AIO2

Official implementation of TGRS paper - AIO2: Online Correction of Object Labels for Deep Learning with Incomplete Annotation in Remote Sensing Image Segmentation

RS-DWL

This is official Pytorch implementation of "Decouple and Weight Semi-supervised Semantic Segmentation of Remote Sensing Images, " ISPRS, 2024.

zhu-xlab.github.io

ChatEarthNet

AdaptMatch

This is official code of "AdaptMatch: Adaptive Matching for Semisupervised Binary Segmentation of Remote Sensing Images," IEEE-TGRS, 2023

SSL4EO-S12

SSL4EO-S12: a large-scale dataset for self-supervised learning in Earth observation

ForestFormer

rrsis

DDM-Former

LCZ_LDL

Official Code for TGRS Publication "Going Beyond One-Hot Encoding: Can Human Uncertainty Improve Model Performance in Earth Observation?" (Koller, Kauermann, Zhu 2023)

Cross-cityDTL

PixelDINO

Semi-supervised RTS segmentation

So2Sat-LCZ42

Zhu et al., So2Sat LCZ42: A Benchmark Dataset for Global Local Climate Zones Classification, IEEE Geoscience and Remote Sensing Magazine, in press, 2020.

HED-UNet

Simultaneous Semantic Segmentation and Edge Detection

PFST

Official implementation of TGRS paper: "Pseudo Features Guided Self-training for Domain Adaptive Semantic Segmentation of Satellite Images"

HTC-DC-Net

Official Implementation of submitted TGRS paper "HTC-DC Net: A Network for Monocular Height Estimation from Single Remote Sensing Images"

FGMAE

Feature guided masked Autoencoder for self-supervised learning in remote sensing

UCDFormer

Code for TGRS paper "UCDFormer: Unsupervised Change Detection Using a Transformer-driven Image Translation"

LULC_MSI_QCNN

Land Cover Classification from Sentinel-2 Images with Quantum-Classical Convolutional Neural Networks (FQCNN and MQCNN)

LRMixer

We propose an efficient post-upsampling architecture, i.e., LRMixer, which aims at yielding a high-resolution HS image on limited hardware resources.

DeCUR

DeCUR: decoupling common & unique representations for multimodal self-supervision.

DINO-MM

Self-supervised vision transformers for joint SAR-optical representation learning

COBRA

Charting Outlines by Recurrent Adaptation – A Deep Contour Detector developed for Charting Greenland's Glacier Calving Fronts.

ML4Earth-Hackathon-2023

Floodcast