Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,64 @@
|
|
1 |
---
|
2 |
license: cc-by-nc-4.0
|
3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
license: cc-by-nc-4.0
|
3 |
---
|
4 |
+
|
5 |
+
# NEO 360: Neural Fields for Sparse View Synthesis of Outdoor Scenes
|
6 |
+
|
7 |
+
[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)<img src="demo/Pytorch_logo.png" width="10%">
|
8 |
+
|
9 |
+
This repository is the pytorch implementation of our paper:
|
10 |
+
<a href="https://www.tri.global/" target="_blank">
|
11 |
+
<img align="right" src="demo/tri-logo.png" width="25%"/>
|
12 |
+
</a>
|
13 |
+
|
14 |
+
**NEO 360: Neural Fields for Sparse View Synthesis of Outdoor Scenes**<br>
|
15 |
+
[__***Muhammad Zubair Irshad***__](https://zubairirshad.com), [Sergey Zakharov](https://zakharos.github.io/), [Katherine Liu](https://www.thekatherineliu.com/), [Vitor Guizilini](https://www.linkedin.com/in/vitorguizilini), [Thomas Kollar](http://www.tkollar.com/site/), [Adrien Gaidon](https://adriengaidon.com/), [Zsolt Kira](https://faculty.cc.gatech.edu/~zk15/), [Rares Ambrus](https://www.tri.global/about-us/dr-rares-ambrus) <br>
|
16 |
+
International Conference on Computer Vision (ICCV), 2023<br>
|
17 |
+
|
18 |
+
[[Project Page](https://zubair-irshad.github.io/projects/neo360.html)] [[arXiv](https://arxiv.org/abs/2308.12967)] [[PDF](https://arxiv.org/pdf/2308.12967.pdf)] [[Video](https://youtu.be/avmylyL_V8c?si=eeTPhl0xJxM3fSF7)]
|
19 |
+
|
20 |
+
|
21 |
+
<p align="center">
|
22 |
+
<img src="demo/NEO_Website_1.jpg" width="100%">
|
23 |
+
</p>
|
24 |
+
|
25 |
+
<p align="center">
|
26 |
+
<img src="demo/NEO_Architecture.JPG" width="100%">
|
27 |
+
</p>
|
28 |
+
|
29 |
+
### Code Coming Soon!
|
30 |
+
|
31 |
+
## 📊 Dataset
|
32 |
+
|
33 |
+
### NERDS 360 Multi-View dataset for Outdoor Scenes
|
34 |
+
|
35 |
+
NeRDS 360: "NeRF for Reconstruction, Decomposition and Scene Synthesis of 360° outdoor scenes” dataset comprising 75 unbounded scenes with full multi-view annotations and diverse scenes for generalizable NeRF training and evaluation.
|
36 |
+
|
37 |
+
<p align="center">
|
38 |
+
<img src="demo/github_dataset.gif" width="100%">
|
39 |
+
</p>
|
40 |
+
|
41 |
+
#### Download the dataset:
|
42 |
+
* [NERDS360 Training Set](https://tri-ml-public.s3.amazonaws.com/github/neo360/datasets/PDMultiObjv6.tar.gz) - 75 Scenes (19.5 GB)
|
43 |
+
* [NERDS360 Test Set](https://tri-ml-public.s3.amazonaws.com/github/neo360/datasets/PD_v6_test.tar.gz) - 5 Scenes (2.1 GB)
|
44 |
+
|
45 |
+
#### Visualizing the dataset (Coming Soon):
|
46 |
+
We will release our visualization scripts to generate visualizations like below i.e. plot accumulated pointclouds, multi-view camera annotations etc.
|
47 |
+
|
48 |
+
<p align="center">
|
49 |
+
<img src="demo/cameras.gif" width="100%">
|
50 |
+
</p>
|
51 |
+
|
52 |
+
## Citation
|
53 |
+
|
54 |
+
If you find this repository or our NERDS 360 dataset useful, please consider citing:
|
55 |
+
|
56 |
+
```
|
57 |
+
@inproceedings{irshad2023neo360,
|
58 |
+
title={NeO 360: Neural Fields for Sparse View Synthesis of Outdoor Scenes},
|
59 |
+
author={Muhammad Zubair Irshad and Sergey Zakharov and Katherine Liu and Vitor Guizilini and Thomas Kollar and Adrien Gaidon and Zsolt Kira and Rares Ambrus},
|
60 |
+
journal={Interntaional Conference on Computer Vision (ICCV)},
|
61 |
+
year={2023},
|
62 |
+
url={https://arxiv.org/abs/2308.12967},
|
63 |
+
}
|
64 |
+
```
|