Training from pre-trained weights (Faster-rcnn/Mask-rcnn) - Object Detection API

I am training Mask-RCNN object detection API from pre-trained weights from the object detection zoo with the coco 2017 dataset.

I have tried both running locally or using the legacy training script. As soon as the training starts, I am never able to see it resume from the mAP mentioned in the Zoo for MaskRCNN or FasterRCNN.

All I want to do is resume training from the mAP of pretrained weights (mentioned 33 for mask_rcnn_resnet101_atrous_coco here at the Zoo)

For instance the very first eval result for masks after 1200 iterations or so during training is given below (using script object_detection/model_main.py from running locally approach).

 Average Precision  (AP) @[ IoU=0.50      | area=   all | maxDets=100 ] = 0.010
 Average Precision  (AP) @[ IoU=0.75      | area=   all | maxDets=100 ] = 0.002
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.000
 Average Precision  (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.001
 Average Precision  (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.010
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=  1 ] = 0.003
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets= 10 ] = 0.005
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=   all | maxDets=100 ] = 0.005
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.000
 Average Recall     (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.002
 Average Recall     (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.013

from Training from pre-trained weights (Faster-rcnn/Mask-rcnn) - Object Detection API