# Targets for multi-task semantic segmentation

**URL:** <https://discuss.pytorch.org/t/targets-for-multi-task-semantic-segmentation/63410>\
**Category:** vision\
**Created:** [December 8, 2019, 6:24pm UTC](https://discuss.pytorch.org/t/targets-for-multi-task-semantic-segmentation/63410 "2019-12-08T18:24:11Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![joekid](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/joekid/32/22490_2.png) [@joekid](https://discuss.pytorch.org/u/joekid)\
**Post date:** [December 8, 2019, 6:24pm UTC](https://discuss.pytorch.org/t/targets-for-multi-task-semantic-segmentation/63410/1 "2019-12-08T18:24:11Z")

</div>

I appeal to your experience, I am newbie in DL.

I like to test the semantic segmentation performance into a multitask, in which, a task specializes on a single label. In this scenario, each task has its own loss, so each task requires its own targets.

**My main question is how to built the targets for each task ?** I am working on `CamVid` dataset (on this, a target annotated image has 12 clases/labels ), Should I post-process each annotated image (if yes, how it would be the class ‘0’ zero, for example)?

All suggestions are welcome.