1. Input Images
  • Upload 2 or more images of the same scene from different views
  • For best results, ensure images have good overlap
  1. Step 1: DUSt3R Initialization & Feature Extraction
  • Click "RUN Step 1" to process your images
  • This step estimates initial DUSt3R point cloud and camera poses, and extracts DUSt3R features for each pixel
  1. Step 2: Readout 3DGS from Features
  • Set the number of training iterations, larger number leads to better quality but longer time (default: 2000, max: 8000)
  • Click "RUN Step 2" to optimize the 3D model
  1. Step 3: Video Rendering
  • Choose a camera trajectory
  • Click "RUN Step 3" to generate a video of your 3D model
  • Processing time depends on image resolution (recommended <1K) and quantity (recommended 2-6 views)
  • For optimal performance, test on high-end GPUs (A100/4090)
  • Use the mouse to interact with 3D models:
    • Left-click: Rotate
    • Scroll: Zoom
    • Right-click: Move
Feat2GS: Probing Visual Foundation Models with Gaussian Splatting

  arxiv    Code  X   Bluesky

  Left-click to rotate, Scroll to zoom, and Right-click to move.
  Failed to visualize or got a GPU runtime error? Run the demo on your local computer!
Camera trajectory

📝 Citation

If you find our work useful for your research or applications, please consider citing the following paper:

@article{chen2024feat2gs,
  title={Feat2GS: Probing Visual Foundation Models with Gaussian Splatting},
  author={Chen, Yue and Chen, Xingyu and Chen, Anpei and Pons-Moll, Gerard and Xiu, Yuliang},
  journal={arXiv preprint arXiv:2412.09606},
  year={2024}
}