Add library_name metadata and link the demo Space

README.md

@@ -6,13 +6,14 @@ tags:
 - video
 - arxiv:2609.30222
 pipeline_tag: keypoint-detection
+library_name: pytorch
 ---
 
 # TrackEverything
 
 ### Long Horizon Dense Tracking via De-Duplicating 3D Scene Representations
 
-[Project Page](https://trackeverything.github.io/) · [Paper](https://arxiv.org/abs/2609.30222) · [Code](https://github.com/ayushjain1144/trackeverything)
+[Project Page](https://trackeverything.github.io/) · [Paper](https://arxiv.org/abs/2609.30222) · [Code](https://github.com/ayushjain1144/trackeverything) · [Demo](https://huggingface.co/spaces/hugging-apps/trackeverything)
 
 TrackEverything tracks dense 3D points across long video sequences. Given RGB frames along with
 depth and camera poses, it builds and maintains a persistent, de-duplicated 3D scene
@@ -26,6 +27,9 @@ This repo holds the trained checkpoint (`model.pth`). Access is gated: request a
 authenticate with `hf auth login` (older `huggingface_hub`: `huggingface-cli login`), or pass
 `token=` explicitly, before downloading.
 
+To try the model on your own video without downloading anything, use the hosted
+[demo Space](https://huggingface.co/spaces/hugging-apps/trackeverything).
+
 ## Installation
 
 Tested on Linux with Python 3.10, PyTorch 2.8, and CUDA 12.9.