Summary of Grin: Zero-shot Metric Depth with Pixel-level Diffusion, by Vitor Guizilini et al.
GRIN: Zero-Shot Metric Depth with Pixel-Level Diffusion
by Vitor Guizilini, Pavel Tokmakov, Achal Dave, Rares Ambrus
First submitted to arxiv on: 15 Sep 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Machine Learning (cs.LG); Image and Video Processing (eess.IV)
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 This paper presents GRIN, an efficient diffusion model for 3D reconstruction from single images, which learns to generate accurate predictions across domains without relying on dense ground-truth labels. By leveraging image features with 3D geometric positional encodings, GRIN conditions the diffusion process both globally and locally, producing depth predictions at a pixel-level. The model is trained using sparse unstructured training data and outperforms state-of-the-art approaches in zero-shot metric monocular depth estimation on eight indoor and outdoor datasets. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper solves a long-standing problem in computer vision: reconstructing 3D images from single pictures. It develops a new way to do this using “diffusion models”, which are good at learning patterns in data. The model, called GRIN, can take in incomplete information and still produce accurate results. The researchers tested GRIN on many different datasets and found that it performed better than other methods. |
Keywords
» Artificial intelligence » Depth estimation » Diffusion » Diffusion model » Zero shot
