Create README.md

README.md

@@ -0,0 +1,17 @@
+The processed ScanNet datasets by VGGT-Det.
+
+If the dataset is helpful, please cite:
+
+@inproceedings{dai2017scannet,
+    title={ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenes},
+    author={Dai, Angela and Chang, Angel X. and Savva, Manolis and Halber, Maciej and Funkhouser, Thomas and Nie{\ss}ner, Matthias},
+    booktitle = {Proc. Computer Vision and Pattern Recognition (CVPR), IEEE},
+    year = {2017}
+}
+
+@inproceedings{cao2026vggtdet,
+  title={VGGT-Det: Mining VGGT Internal Priors for Sensor-Geometry-Free Multi-View Indoor 3D Object Detection},
+  author={Cao, Yang and Wu, Feize and Dave Chen, Zhenyu and Zhong, Yingji and Hong, Lanqing and Xu, Dan},
+  booktitle={CVPR},
+  year={2026}
+}