Summary of Fawn: Floor-and-walls Normal Regularization For Direct Neural Tsdf Reconstruction, by Anna Sokolova et al.
FAWN: Floor-And-Walls Normal Regularization for Direct Neural TSDF Reconstruction
by Anna Sokolova, Anna Vorontsova, Bulat Gabdullin, Alexander Limonov
First submitted to arxiv on: 17 Jun 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Machine Learning (cs.LG)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary A machine learning method for correcting distorted room shapes and eliminating local artifacts in direct 3D reconstruction is proposed. The approach, called FAWN, modifies truncated signed distance function (TSDF) reconstruction methods by detecting walls and floors in a scene and penalizing surface normals that deviate from the horizontal and vertical directions. Implemented as a 3D sparse convolutional module, FAWN can be incorporated into any trainable pipeline predicting TSDF. The method is shown to use semantics more effectively than existing approaches, leading to quality gains in SCANNET, ICL-NUIM, TUM RGB-D, and 7SCENES benchmarks. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary A new way to make 3D reconstruction better is being developed. Right now, it’s hard to get accurate pictures of rooms because the floors and walls can be distorted or have holes in them. The solution proposed in this paper is called FAWN. It looks at the room and tries to correct any mistakes by making sure the surfaces are lined up with the floor and walls. This new method works well with existing technology and makes it easier to get accurate pictures of rooms. |
Keywords
» Artificial intelligence » Machine learning » Semantics
