Awesome

DINO-MM

Self-supervised vision transformers for joint SAR-optical representation learning, IGARSS 2022

Note: Codes may not run out-of-the-box, a cleaner version to be organized.

training and linear evaluation

run sbatch scripts/slurm/sar_optical/srun_xxx.sh

pre-trained model

DINO-MM with ViT-S/8, input 14 bands: checkpoint, training log

Further reads

"Self-supervised Learning in Remote Sensing: A Review". paper | Repository

"SSL4EO-S12: A Large-Scale Multi-Modal, Multi-Temporal Dataset for Self-Supervised Learning in Earth Observation". Paper | Repository

"DeCUR: decoupling common & unique representations for multimodal self-supervision". paper | Repository