Add task category and dataset description

README.md

@@ -1,12 +1,24 @@
 ---
 license: cc-by-nc-4.0
+task_categories:
+- other
+tags:
+- 3d
+- gaussian-splatting
+- quality-assessment
 ---
+
 # MUGSQA
 
 <a href="https://arxiv.org/abs/2511.06830"><img src="https://img.shields.io/badge/Paper-b5212f.svg?logo=arxiv" alt="arXiv" style="vertical-align: middle;"></a>
 <a href="https://github.com/Solivition/MUGSQA"><img src="https://img.shields.io/badge/GitHub-%23121011.svg?logo=github" alt="github" style="vertical-align: middle;"></a>
 
-Officially released MUGSQA dataset for the <strong>ICASSP 2026</strong> paper <strong>MUGSQA: Novel Multi-Uncertainty-Based Gaussian Splatting Quality Assessment Method, Dataset, and Benchmarks</strong>.
+Officially released MUGSQA dataset for the **ICASSP 2026** paper **MUGSQA: Novel Multi-Uncertainty-Based Gaussian Splatting Quality Assessment Method, Dataset, and Benchmarks**.
+
+MUGSQA (Multi-Uncertainty-Based Gaussian Splatting Quality Assessment) is a novel dataset designed to evaluate the perceptual quality of 3D objects reconstructed using Gaussian Splatting (GS). The dataset is constructed considering multiple uncertainties of the input data, including:
+- The quantity and resolution of input views.
+- The view distance.
+- The accuracy of the initial point cloud.
 
 We are currently uploading the dataset, and instructions for using it will be added soon.