Summary of 360 in the Wild: Dataset For Depth Prediction and View Synthesis, by Kibaek Park et al.
360 in the Wild: Dataset for Depth Prediction and View Synthesis
by Kibaek Park, Francois Rameau, Jaesik Park, In So Kweon
First submitted to arxiv on: 27 Jun 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Artificial Intelligence (cs.AI)
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 The paper introduces a large-scale dataset of panoramic videos captured in various locations worldwide, featuring diverse environments and contexts. The dataset contains 25K images with corresponding camera poses and depth maps. It demonstrates its relevance by applying it to two tasks: single-image depth estimation and view synthesis. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper creates a huge database of 360-degree videos taken from different places around the world. The videos are mixed, showing both indoor and outdoor scenes, and sometimes moving objects. The dataset has 25,000 images with information about where the camera was and how far away things were. It’s useful for two important tasks: guessing distances in one image and creating new views. |
Keywords
» Artificial intelligence » Depth estimation
