# \#feature-request

**URL:** https://community.labelstud.io/tag/feature-request/17.md

[Latest](https://community.labelstud.io/latest.md) · [Categories](https://community.labelstud.io/categories.md) · [Tags](https://community.labelstud.io/tags.md)

---

## [Fixed RectangleLabel aspect ratios](https://community.labelstud.io/t/fixed-rectanglelabel-aspect-ratios/391)

<div class="topic-metadata">

**Author:** [@orfeo](https://community.labelstud.io/u/orfeo)\
**Replies:** 0\
**Last updated:** [October 23, 2024, 1:58pm UTC](https://community.labelstud.io/t/fixed-rectanglelabel-aspect-ratios/391 "2024-10-23T13:58:18Z")

</div>

For text2image LORA-training, it is beneficial to have images with equal dimensions, because multiple images can be combined in one batch. Scaling down images is no problem, but changing aspect ratio is! I would therefor…

---

## [Support for Polygon Labels in Video Object Detection and Tracking](https://community.labelstud.io/t/support-for-polygon-labels-in-video-object-detection-and-tracking/163)

<div class="topic-metadata">

**Author:** [@sajarin](https://community.labelstud.io/u/sajarin)\
**Replies:** 1\
**Last updated:** [May 10, 2024, 5:37pm UTC](https://community.labelstud.io/t/support-for-polygon-labels-in-video-object-detection-and-tracking/163 "2024-05-10T17:37:04Z")

</div>

Question: Can Polygon labels be used for Video Object Detection and Tracking template, or is it only possible with VideoRectangle, aka bounding boxes? Answer Unfortunately, as of now, polygon labels are not supported …