thanks for this excellent approach
but when i runned "make sh-gt.py"
i got error：Failed to interpret file 'XX.npy' as a pickle
even i have setted "allow_pickle=True" or numpy==1.14.1 in paper "https://github.com/StevenLiuWen/ano_pred_cvpr2018"
can you share the “fram_test_mask” you used when generating the gt
thank you~
thanks for this excellent approach but when i runned "make sh-gt.py" i got error：Failed to interpret file 'XX.npy' as a pickle even i have setted "allow_pickle=True" or numpy==1.14.1 in paper "https://github.com/StevenLiuWen/ano_pred_cvpr2018" can you share the “fram_test_mask” you used when generating the gt thank you~