MetaDent AppImage annotation app

There are limited free open-source tools for labeling images with multi-modal annotations (e.g., free-text labels + segmentation masks) which are essential for training/evaluating vision-language models.

This app provides a simple interface for annotating images with:

1. An overall description of the image,
2. A list of free-text labels, each associated with segmentation(s).

It also supports plugin-server, allowing integration with external models for auto-labeling.

Paper & Citation ​

MetaDent is a intra-oral image dataset, benchmark, and an annotation app for vision-language tasks in dentistry. This application is part of the MetaDent project.

MetaDent: Labeling Clinical Images for Vision-Language Models in Dentistry
Meng-Xun Li†, Wen-Hui Deng†, Zhi-Xing Wu, Chun-Xiao Jin, Jia-Min Wu, Yue Han, James Kit Hon Tsoi, Gui-Song Xia*, Cui Huang*
Journal of Dental Research, 2026.
[Paper Link] [Project Page]