Detection_eval of shared pretrained model is very less

[manogna-s]

I evaluated the pretrained model 'model_icdar15.caffemodel' shared on ICDAR15 test set with 500 images . The detection_eval is observed to be:

Resolution: 768x768 , IoU threshold: 0.5, detection_eval=0.463 Resolution: 1024x1024 , IoU threshold: 0.5, detection_eval=0.469

However, in the paper the precision is mentioned to be 0.872 for 1024x1024 while I get 0.469. The metrics used are Pascal VOC based 11-point method to get mAP right?

[github14343]

Have you solved the problem? Thanks!