How to reach the mAP on Maskrcnn paper?

[yiyang186]

I used the project to train on coco train 2014. And I got the evaluation:

loading annotations into memory...
Done (t=4.37s)
creating index...
index created!
40504
40504
Loading and preparing results...
DONE (t=11.58s)
creating index...
index created!
Running per image evaluation...
Evaluate annotation type keypoints
DONE (t=147.12s).
Accumulating evaluation results...
DONE (t=1.72s).
Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets= 20 ] = 0.229
Average Precision (AP) @[ IoU=0.50 | area= all | maxDets= 20 ] = 0.614
Average Precision (AP) @[ IoU=0.75 | area= all | maxDets= 20 ] = 0.116
Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets= 20 ] = 0.191
Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets= 20 ] = 0.281
Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 20 ] = 0.310
Average Recall (AR) @[ IoU=0.50 | area= all | maxDets= 20 ] = 0.686
Average Recall (AR) @[ IoU=0.75 | area= all | maxDets= 20 ] = 0.241
Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets= 20 ] = 0.259
Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets= 20 ] = 0.378

How do I train to mAP(kp, 50)=80+ on maskrcnn paper?

[chenyanyin]

i want to know,too