README.md

---
pipeline_tag: feature-extraction
---

<!-- # NeRF-MAE: Masked AutoEncoders for Self-Supervised 3D Representation Learning for Neural Radiance Fields -->

<div align="center">
    <img src="demo/nerf-mae_teaser.png" width="85%">
    <img src="demo/nerf-mae_teaser.jpeg" width="85%">
</div>
<!-- <p align="center">
<img src="demo/nerf-mae_teaser.jpeg" width="100%">
</p> -->

<br>
<div align="center">

[![arXiv](https://img.shields.io/badge/arXiv-2404.01300-gray?style=for-the-badge&logo=arxiv&logoColor=white&color=B31B1B)](https://arxiv.org/abs/2404.01300)
[![Project Page](https://img.shields.io/badge/Project-Page-orange?style=for-the-badge&logoColor=white&labelColor=gray&link=https%3A%2F%2Fnerf-mae.github.io%2F)](https://nerf-mae.github.io)
[![Pytorch](https://img.shields.io/badge/Pytorch-%3E1.12-gray?style=for-the-badge&logo=pytorch&logoColor=white&labelColor=gray&color=ee4c2c&link=https%3A%2F%2Fnerf-mae.github.io%2F)](https://pytorch.org/)
[![Cite](https://img.shields.io/badge/Cite-Bibtex-gray?style=for-the-badge&logoColor=white&color=F7A41D
)](https://github.com/zubair-irshad/NeRF-MAE?tab=readme-ov-file#citation)
[![Video](https://img.shields.io/badge/youtube-video-CD201F?style=for-the-badge&logo=youtube&labelColor=grey
)](https://youtu.be/D60hlhmeuJI?si=d4RfHAwBJgLJXdKj)



</div>

---

<a href="https://www.tri.global/" target="_blank">
 <img align="right" src="demo/GeorgiaTech_RGB.png" width="18%"/>
</a>

<a href="https://www.tri.global/" target="_blank">
 <img align="right" src="demo/tri-logo.png" width="17%"/>
</a>

### [Project Page](https://nerf-mae.github.io/) | [arXiv](https://arxiv.org/abs/2308.12967) | [PDF](https://arxiv.org/pdf/2308.12967.pdf)



**NeRF-MAE : Masked AutoEncoders for Self-Supervised 3D Representation Learning for Neural Radiance Fields**

<a href="https://zubairirshad.com"><strong>Muhammad Zubair Irshad</strong></a>
·
<a href="https://zakharos.github.io/"><strong>Sergey Zakharov</strong></a>
·
<a href="https://www.linkedin.com/in/vitorguizilini"><strong>Vitor Guizilini</strong></a>
·
<a href="https://adriengaidon.com/"><strong>Adrien Gaidon</strong></a>
·
<a href="https://faculty.cc.gatech.edu/~zk15/"><strong>Zsolt Kira</strong></a>
·
<a href="https://www.tri.global/about-us/dr-rares-ambrus"><strong>Rares Ambrus</strong></a>
<br> **European Conference on Computer Vision, ECCV 2024**<br>

<b> Toyota Research Institute &nbsp; | &nbsp; Georgia Institute of Technology</b>

## 💡 Highlights
- **NeRF-MAE**: The first large-scale pretraining utilizing Neural Radiance Fields (NeRF) as an input modality. We pretrain a single Transformer model on thousands of NeRFs for 3D representation learning.
- **NeRF-MAE Dataset**: A large-scale NeRF pretraining and downstream task finetuning dataset.

## 🏷️ TODO 🚀

- [x] Release large-scale pretraining code 🚀
- [x] Release NeRF-MAE dataset comprising radiance and density grids 🚀
- [x] Release 3D object detection finetuning and eval code 🚀
- [x] Pretrained NeRF-MAE checkpoints and out-of-the-box model usage 🚀

## NeRF-MAE Model Architecture
<p align="center">
<img src="demo/nerf-mae_architecture.jpg" width="90%">
</p>


## Citation

If you find this repository or our dataset useful, please star ⭐ this repository and consider citing 📝:

```
@inproceedings{irshad2024nerfmae,
    title={NeRF-MAE: Masked AutoEncoders for Self-Supervised 3D Representation Learning for Neural Radiance Fields},
    author={Muhammad Zubair Irshad and Sergey Zakharov and Vitor Guizilini and Adrien Gaidon and Zsolt Kira and Rares Ambrus},
    booktitle={European Conference on Computer Vision (ECCV)},
    year={2024}
    }
```

### Contents
 - [🌇  Environment](#-environment)
 - [⛳ Model Usage and Checkpoints](#-model-usage-and-checkpoints)
 - [🗂️ Dataset](#-dataset)

 ## 🌇  Environment

Create a python 3.7 virtual environment and install requirements:

```bash
cd $NeRF-MAE repo
conda create -n nerf-mae python=3.9
conda activate nerf-mae
pip install --upgrade pip
pip install -r requirements.txt
pip install torch==1.12.1+cu113 torchvision==0.13.1+cu113 -f https://download.pytorch.org/whl/torch_stable.html
```
The code was built and tested on **cuda 11.3**

Compile CUDA extension, to run downstream task finetuning, as described in [NeRF-RPN](https://github.com/lyclyc52/NeRF_RPN):

```bash
cd $NeRF-MAE repo
cd nerf_rpn/model/rotated_iou/cuda_op
python setup.py install
cd ../../../..

```

## ⛳ Model Usage and Checkpoints

- [Hugginface repo to download pretrained and finetuned checkpoints](https://huggingface.co/mirshad7/NeRF-MAE)

NeRF-MAE is structured to provide easy access to pretrained NeRF-MAE models (and reproductions), to facilitate use for various downstream tasks. This is for extracting good visual features from NeRFs if you don't have resources for large-scale pretraining. Our pretraining provides an easy-to-access embedding of any NeRF scene, which can be used for a variety of downstream tasks in a straightforwaed way. 

We have released pretrained and finetuned checkpoints to start using our codebase out-of-the-box. There are two types of usages. 1. Most common one is using the features directly in a downstream task such as an FPN head for 3D Object Detection and 2. Reconstruct the original grid for enforcing losses such as masked reconstruction loss. Below is a sample useage of our model with spelled out comments.


1. Get the features to be used in a downstream task

```python
import torch

# Define Swin Transformer configurations
swin_config = {
    "swin_t": {"embed_dim": 96, "depths": [2, 2, 6, 2], "num_heads": [3, 6, 12, 24]},
    "swin_s": {"embed_dim": 96, "depths": [2, 2, 18, 2], "num_heads": [3, 6, 12, 24]},
    "swin_b": {"embed_dim": 128, "depths": [2, 2, 18, 2], "num_heads": [3, 6, 12, 24]},
    "swin_l": {"embed_dim": 192, "depths": [2, 2, 18, 2], "num_heads": [6, 12, 24, 48]},
}

# Set the desired backbone type
backbone_type = "swin_s"
config = swin_config[backbone_type]

# Initialize Swin Transformer model
model = SwinTransformer_MAE3D_New(
    patch_size=[4, 4, 4],
    embed_dim=config["embed_dim"],
    depths=config["depths"],
    num_heads=config["num_heads"],
    window_size=[4, 4, 4],
    stochastic_depth_prob=0.1,
    expand_dim=True,
    resolution=resolution,
)

# Load checkpoint and remove unused layers
checkpoint_path = hf_hub_download(repo_id="mirshad7/NeRF-MAE", filename="nerf_mae_pretrained.pt")
checkpoint = torch.load(checkpoint_path, map_location="cpu")
model.load_state_dict(checkpoint["state_dict"])
for attr in ["decoder4", "decoder3", "decoder2", "decoder1", "out", "mask_token"]:
    delattr(model, attr)

# Extract features using Swin Transformer backbone. input_grid has sample shape torch.randn((1, 4, 160, 160, 160))
features = []
input_grid = model.patch_partition(input_grid) + model.pos_embed.type_as(input_grid).to(input_grid.device).clone().detach()
for stage in model.stages:
    input_grid = stage(input_grid)
    features.append(torch.permute(input_grid, [0, 4, 1, 2, 3]).contiguous())  # Format: [N, C, H, W, D]

#Multi-scale features have shape:  [torch.Size([1, 96, 40, 40, 40]), torch.Size([1, 192, 20, 20, 20]), torch.Size([1, 384, 10, 10, 10]), torch.Size([1, 768, 5, 5, 5])] 

# Process features through FPN
```

2. Get the Original Grid Output 
```python
import torch
# Load data from the specified folder and filename with the given resolution.
res, rgbsigma = load_data(folder_name, filename, resolution=args.resolution)

# rgbsigma has sample shape torch.randn((1, 4, 160, 160, 160))

# Build the model using provided arguments.
model = build_model(args)

# Load checkpoint if provided.
if args.checkpoint:
    model.load_state_dict(torch.load(args.checkpoint, map_location="cpu")["state_dict"])
    model.eval()  # Set model to evaluation mode.

# Run inference getting the features out for downsteam usage
with torch.no_grad():
    pred = model([rgbsigma], is_eval=True)[3]  # Extract only predictions.

```

### 1. How to plug these features for downstream 3D bounding detection from NeRFs (i.e. plug-and-play with a [NeRF-RPN](https://github.com/lyclyc52/NeRF_RPN) OBB prediction head)

Please also see the section on [Finetuning](#-finetuning). Our released finetuned checkpoint achieves state-of-the-art on 3D object detection in NeRFs. To run evaluation using our finetuned checkpoint on the dataset provided by NeRF-RPN, please run the below script, after updating the paths to the pretrained checkpoint i.e. --checkpoint and  DATA_ROOT depending on evaluation done for ```Front3D``` or ```Scannet```:

```
bash test_fcos_pretrained.sh
```

Also see the cooresponding run file i.e. ```run_fcos_pretrained.py``` and our model adaptation i.e. ```SwinTransformer_FPN_Pretrained_Skip```. This is a minimal adaptation to plug and play our weights with a NeRF-RPN architecture and achieve significant boost in performance. 


## 🗂️ Dataset

Download the preprocessed datasets here. 

- Pretraining dataset (comprising NeRF radiance and density grids). [Download link](https://s3.amazonaws.com/tri-ml-public.s3.amazonaws.com/github/nerfmae/NeRF-MAE_pretrain.tar.gz)
- Finetuning dataset (comprising NeRF radiance and density grids and bounding box/semantic labelling annotations). [3D Object Detection (Provided by NeRF-RPN)](https://drive.google.com/drive/folders/1q2wwLi6tSXu1hbEkMyfAKKdEEGQKT6pj), [3D Semantic Segmentation (Coming Soon)](), [Voxel-Super Resolution (Coming Soon)]()


Extract pretraining and finetuning dataset under ```NeRF-MAE/datasets```. The directory structure should look like this:

```
NeRF-MAE
├── pretrain
│   ├── features
│   └── nerfmae_split.npz
└── finetune
    └── front3d_rpn_data
        ├── features
        ├── aabb
        └── obb
```


Note: The above datasets are all you need to train and evaluate our method. Bonus: we will be releasing our multi-view rendered posed RGB images from FRONT3D, HM3D and Hypersim as well as Instant-NGP trained checkpoints soon (these comprise over 1M+ images and 3k+ NeRF checkpoints)

Please note that our dataset was generated using the instruction from [NeRF-RPN](https://github.com/lyclyc52/NeRF_RPN) and [3D-CLR](https://vis-www.cs.umass.edu/3d-clr/). Please consider citing our work, NeRF-RPN and 3D-CLR if you find this dataset useful in your research. 

Please also note that our dataset uses [Front3D](https://arxiv.org/abs/2011.09127), [Habitat-Matterport3D](https://arxiv.org/abs/2109.08238), [HyperSim](https://github.com/apple/ml-hypersim) and [ScanNet](https://www.scan-net.org/) as the base version of the dataset i.e. we train a NeRF per scene and extract radiance and desnity grid as well as aligned NeRF-grid 3D annotations. Please read the term of use for each dataset if you want to utilize the posed multi-view images for each of these datasets. 

### For More details, please checkout out Paper, Github and Project Page!

---
license: cc-by-nc-4.0
---