Putting NeRF on a Diet: Semantically Consistent Few-Shot View Synthesis

Last update: Dec 27, 2022

Related tags

Overview

Putting NeRF on a Diet: Semantically Consistent Few-Shot View Synthesis

This repository contains the official implementation of DietNeRF, a system that reconstructs 3D scenes from a few posed photos.

Setup

We use the following folder structure:

dietnerf/
  logs/ (images, videos, checkpoints)
  data/
    nerf_synthetic/
  configs/ (run configuration files)
CLIP/ (Fork of OpenAI's clip repository with a wrapper)

Create conda environment:

conda create -n dietnerf python=3.9
conda activate dietnerf

Set up requirements and our fork of CLIP:

pip install -r requirements.txt
cd CLIP
pip install -e .

wandb login

Experiments on the Realistic Synthetic dataset

Realistic Synthetic experiments are implemented in the ./dietnerf subdirectory.

You need to download datasets from NeRF's Google Drive folder. The dataset was used in the original NeRF paper by Mildenhall et al. For example,

mkdir dietnerf/logs/ dietnerf/data/
cd dietnerf/data
pip install gdown
gdown --id 18JxhpWD-4ZmuFKLzKlAw-w5PpzZxXOcG -O nerf_synthetic.zip
unzip nerf_synthetic.zip
rm -r __MACOSX

Then, shrink images to 400x400:

python dietnerf/scripts/bulk_shrink_images.py "dietnerf/data/nerf_synthetic/*/*/*.png" dietnerf/data/nerf_synthetic_400_rgb/ True

These images are used for FID/KID computation. The dietnerf/run_nerf.py training and evaluation code automatically shrinks images with the --half_res argument.

Each experiment has a config file stored in dietnerf/configs/. Scripts in dietnerf/scripts/ can be run to train and evaluate models. Run these scripts from ./dietnerf. The scripts assume you are running one script at a time on a server with 8 NVIDIA GPUs.

cd dietnerf
export WANDB_ENTITY=
   
    

# NeRF baselines
sh scripts/run_synthetic_nerf_100v.sh
sh scripts/run_synthetic_nerf_8v.sh
sh scripts/run_synthetic_simplified_nerf_8v.sh

# DietNeRF with 8 observed views
sh scripts/run_synthetic_dietnerf_8v.sh
sh scripts/run_synthetic_dietnerf_ft_8v.sh

# NeRF and DietNeRF with partial observability
sh scripts/run_synthetic_unseen_side_14v.sh

Experiments on the DTU dataset

Coming soon. Our paper also fine-tunes pixelNeRF on DTU scenes for 1-shot view synthesis.

Citation and acknowledgements

If DietNeRF is relevant to your project, please cite our associated paper:

@InProceedings{Jain_2021_ICCV,
    author    = {Jain, Ajay and Tancik, Matthew and Abbeel, Pieter},
    title     = {Putting NeRF on a Diet: Semantically Consistent Few-Shot View Synthesis},
    booktitle = {Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV)},
    month     = {October},
    year      = {2021},
    pages     = {5885-5894}
}

This code is based on Yen-Chen Lin's PyTorch implementation of NeRF and the official pixelNeRF code.

Putting NeRF on a Diet: Semantically Consistent Few-Shot View Synthesis

Related tags

Overview

Putting NeRF on a Diet: Semantically Consistent Few-Shot View Synthesis

Setup

Experiments on the Realistic Synthetic dataset

Experiments on the DTU dataset

Citation and acknowledgements

Owner

Ajay Jain

Code for DeepXML: A Deep Extreme Multi-Label Learning Framework Applied to Short Text Documents

Multimodal commodity image retrieval 多模态商品图像检索

Team Enigma at ArgMining 2021 Shared Task: Leveraging Pretrained Language Models for Key Point Matching

Tensorflow implementation of Human-Level Control through Deep Reinforcement Learning

OpenDILab Multi-Agent Environment

Rule Based Classification Project For Python

The MATH Dataset

This is an official pytorch implementation of Fast Fourier Convolution.

SEC'21: Sparse Bitmap Compression for Memory-Efficient Training onthe Edge

Official PyTorch implementation of Learning Intra-Batch Connections for Deep Metric Learning (ICML 2021) published at International Conference on Machine Learning

UltraGCN: An Ultra Simplification of Graph Convolutional Networks for Recommendation

OpenPCDet Toolbox for LiDAR-based 3D Object Detection.

JAX-based neural network library

Experiment about Deep Person Re-identification with EfficientNet-v2

[ICLR 2021] Heteroskedastic and Imbalanced Deep Learning with Adaptive Regularization

VIsually-Pivoted Audio and(N) Text

General purpose Slater-Koster tight-binding code for electronic structure calculations

FastCover: A Self-Supervised Learning Framework for Multi-Hop Influence Maximization in Social Networks by Anonymous.

Implementation based on Paper - Learning a Probabilistic Latent Space of Object Shapes via 3D Generative-Adversarial Modeling

Source code for Zalo AI 2021 submission