# Mask-RCNN move to java

**URL:** <https://community.konduit.ai/t/mask-rcnn-move-to-java/1170>\
**Category:** DL4J\
**Created:** [February 8, 2021, 6:52pm UTC](https://community.konduit.ai/t/mask-rcnn-move-to-java/1170 "2021-02-08T18:52:49Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![DinisDimitris](https://yyz1.discourse-cdn.com/flex035/user_avatar/community.konduit.ai/dinisdimitris/32/456_2.png) [@DinisDimitris](https://community.konduit.ai/u/DinisDimitris)\
**Post date:** [February 8, 2021, 6:52pm UTC](https://community.konduit.ai/t/mask-rcnn-move-to-java/1170/1 "2021-02-08T18:52:49Z")

</div>

Hi,

We are required to build a project with my team that requires instance segmentation. We are not too familiar with CNN’s however we got some basics. We have managed to train a model using the matterport Mask RCNN ( [GitHub - matterport/Mask\_RCNN: Mask R-CNN for object detection and instance segmentation on Keras and TensorFlow](https://github.com/matterport/Mask_RCNN) ) using keras and tensorflow, and we would need to import it to java. I am thinking of importing the keras weights using the keras model import but I am not sure if that will work? Do you have any advice?

Thank you.