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

Learning Multiresolution Matrix Factorization and its Wavelet Networks on Graphs

OBBDetection is a oriented object detection library, which is based on MMdetection.

Simple (but Strong) Baselines for POMDPs

Object Detection with YOLOv3

Semantic code search implementation using Tensorflow framework and the source code data from the CodeSearchNet project

Efficient Two-Step Networks for Temporal Action Segmentation (Neurocomputing 2021)

Official repository of the paper Privacy-friendly Synthetic Data for the Development of Face Morphing Attack Detectors

Weakly Supervised Posture Mining with Reverse Cross-entropy for Fine-grained Classification

Analysis of Antarctica sequencing samples contaminated with SARS-CoV-2

In the AI for TSP competition we try to solve optimization problems using machine learning.

Implementation of Lie Transformer, Equivariant Self-Attention, in Pytorch

Commonsense Ability Tests

GPOEO is a micro-intrusive GPU online energy optimization framework for iterative applications

Amazon Forest Computer Vision: Satellite Image tagging code using PyTorch / Keras with lots of PyTorch tricks

Official Pytorch implementation of the paper "MotionCLIP: Exposing Human Motion Generation to CLIP Space"

Data and codes for ACL 2021 paper: Towards Emotional Support Dialog Systems

EdiBERT, a generative model for image editing

Official PyTorch implementation of the paper: Improving Graph Neural Network Expressivity via Subgraph Isomorphism Counting.

torchbearer: A model fitting library for PyTorch

AntiFuzz: Impeding Fuzzing Audits of Binary Executables