Mask_rcnn: Can this implementation achieve the state-of-art performance?

Created on 5 Apr 2019  路  5Comments  路  Source: matterport/Mask_RCNN

Hi Guys,

I tested the implementation on the COCO dataset with the pre-trained weights, while I cannot get the state of art performance shown at http://cocodataset.org/#detection-leaderboard. At this point, I am trying to figure out what is going wrong.. Is there any one who can achieve the expected performance with the implementation?

Most helpful comment

They have their own documentation 1 2 about adding datasets.

All 5 comments

It cannot. See #1.

@ppwwyyxx interesting. I guess you're using another implementation, then. May I ask which one?

I use my own implementation in TensorFlow, and maskrcnn-benchmark in PyTorch as well.

Not saying that they reach best performance on COCO leaderboard, but at least they are able to reproduce performance in the official implementation in Detectron.

Nice! Is it as easy to modify your implementation (or maskrcnn-benchmark) to train on a custom dataset, as it is for this one?

They have their own documentation 1 2 about adding datasets.

Was this page helpful?
0 / 5 - 0 ratings