Author the PianoVAM organization card

README.md

@@ -1,10 +1,34 @@
 ---
 title: README
-emoji: 🏢
-colorFrom: blue
-colorTo: blue
+emoji: 🎹
+colorFrom: indigo
+colorTo: purple
 sdk: static
 pinned: false
 ---
 
-Edit this `README.md` markdown file to author your organization card.
+<div align="center">
+
+# PianoVAM 🎹
+
+**A multimodal piano performance dataset, and the models built on it.**
+
+From the [MAC Lab](https://mac.kaist.ac.kr/) at KAIST and the [Music Informatics Group](https://musicinformatics.gatech.edu/) at Georgia Tech.
+
+</div>
+
+PianoVAM captures piano performance across every modality at once. Each of 107 practice recordings on a Yamaha Disklavier aligns top-view video, lossless audio, ground-truth MIDI, and 21-point hand landmarks to the same take. On top of the dataset we build models that read one modality from another, starting with V2N, which transcribes piano to complete MIDI from silent video alone.
+
+### PianoVAM: a multimodal piano dataset (ISMIR 2025)
+- 🗂️ [Dataset](https://huggingface.co/datasets/PianoVAM/PianoVAM_v1)
+- 🎬 [Live demo](https://huggingface.co/spaces/PianoVAM/PianoVAM)
+- 📄 [Paper](https://arxiv.org/abs/2509.08800)
+- 💻 [Code](https://github.com/yonghyunk1m/PianoVAM-Code)
+
+### V2N: visual piano transcription (ISMIR 2026)
+- 🎬 [Live demo](https://huggingface.co/spaces/PianoVAM/V2N)
+- 📄 [Paper](https://arxiv.org/abs/2608.03419)
+- 💻 [Code](https://github.com/yonghyunk1m/V2N)
+
+### PiaRec & ASDF: acquisition and fingering toolkits (ISMIR 2025 LBD)
+- 📄 [Paper](https://arxiv.org/abs/2509.15222)