# ThiefCloud on T-CLOUD ThiefCloud (TCSVT'25) trained on the T-CLOUD training split, as used in the comparison tables of GeoCR ([arXiv:2609.32510](https://arxiv.org/abs/2609.32510)). One checkpoint, seed 0. | File | Bytes | SHA-256 | |---|---|---| | `final.pth` | 8,460,367 | `d83fc785a00fee2e87b75e81c0ce15a8f3046bf78086b57bccd0522d661e9d56` | ## Code - Upstream: at commit `17962e27dff59dacdf5459a0b355d53134b1ae5d`. - Patch, configuration (`configs/tcloud.yaml`) and scripts: [`baselines/thiefcloud`](https://github.com/KAIST-VICLab/GeoCR/tree/main/baselines/thiefcloud) in the GeoCR repository. - Upstream licence: no licence file; the upstream README states "The code can be used for academic purposes only.". ```bash hf download JeonghyeokDo/GeoCR --include "baselines/t-cloud/thiefcloud/*" --local-dir weights REPO=/path/to/patched/ThiefCloud CR_DATASETS=/path/to/datasets \ CKPT=weights/baselines/t-cloud/thiefcloud/final.pth bash GeoCR/baselines/thiefcloud/infer.sh t-cloud ``` ## Citation ```bibtex @article{zhao2025thiefcloud, title = {ThiefCloud: A Thickness Fused Thin Cloud Removal Network for Optical Remote Sensing Image With Self-Supervised Learnable Cloud Prior}, author = {Zhao, Anqi and Feng, Ruitao and Li, Xinghua}, journal = {IEEE Transactions on Circuits and Systems for Video Technology}, volume = {35}, number = {12}, pages = {11834--11848}, year = {2025} } ```