The engine file trained and converted by TLT runs incorrectly in Deepstream on the jetson platform

I use TLT to train Mask-Rcnn, the data set is coco2017, and the errors running in Deepstream are as follows:

any idea?thanks.

Where did you run inference, in Jetson devices?

yep,xavier nx

How did you generate the trt engine, in host PC?

Oh, do I need to run tlt-converter in xavier to get engine files?

Yes, please use tlt-converter jetson version to generate trt engine in your nx.

hi,Morganh i run tlt-converter in xavier to get engine files,It works normally, but the FPS is only 3. What is the problem?

Refer to Training Instance Segmentation Models Using Mask R-CNN on the NVIDIA Transfer Learning Toolkit | NVIDIA Developer Blog
I’m afraid it is similar to your result. (Please run it with batch size 2).

Currently, the Maskrcnn does not support pruning yet. So, if you want to get a higher fps, you can train a smaller network.


  • Input size : C * W * H (where C = 3, W > =128, H >=128 and W, H are multiples of 32)
  • Image format : JPG
  • Label format : COCO detection