2019遥感图像稀疏表征与智能分析竞赛第三名方案

• Mask RCNN 0.3541
• Hybrid Task RCNN + deform conv 0.36633
• expand bbox 0.364
• cascade score thresh adopt to 0.5 0.366
• small number class augmentation 0.372
• cross_entropy weighted 0.369
• sync BN 0.376
• IOU sampler 0.383
• pesudo label fine tune 0.362
• balanced sampler 0.369
• augmentation 0.40
• 3 scale test 0.399
• resnext101 0.41
• scale2 finetune 0.43

Please refer to INSTALL.md for installation and dataset preparation.

python ./tools/prepare_data.py OR Dataprepare.ipynb

data will generate in ./data/rscup/annotation/ and ./data/rscup/train

./tools/dist_train.sh ./configs/rscup/htc_next_3s.py <gpu_num>

./tools/dist_test.sh ./configs/rscup/htc_next_3s.py ./work_dirs/htc_next_3s/epoch*.pth <gpu_num> \

--out test.pkl

python ./tools/merge_result.py

We achieved the converter from Hybrid Task cascade RCNN trained with mmdetection to Caffe.

Please refer " ".