What are the metrics on COCO dataset?

[anuar12]

Would you be great if you could share the metrics on COCO.
Thank you!

[anuar12]

I evaluated Densenet-121 from the provided model and got:

 Average Precision  (AP) @[ IoU=0.50:0.95 | area=   all | maxDets= 20 ] = 0.292
 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets= 20 ] = 0.551
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets= 20 ] = 0.274
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets= 20 ] = 0.213
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets= 20 ] = 0.412
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 20 ] = 0.364
 Average Recall     (AR) @[ IoU=0.50      | area=   all | maxDets= 20 ] = 0.593
 Average Recall     (AR) @[ IoU=0.75      | area=   all | maxDets= 20 ] = 0.363
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets= 20 ] = 0.227
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets= 20 ] = 0.553

which is lower than expected (as CMU-Pose was geeting 0.6+ on mAP)
Would you suggest what improvements I should make to increase mAP?
One thing that I noticed is that I can play with ParseObjects parameters too.