Pytorch implementation for A-NeRF: Articulated Neural Radiance Fields for Learning Human Shape, Appearance, and Pose

Related tags

Deep LearningA-NeRF
Overview

A-NeRF: Articulated Neural Radiance Fields for Learning Human Shape, Appearance, and Pose

Paper | Website | Data

A-NeRF: Articulated Neural Radiance Fields for Learning Human Shape, Appearance, and Pose
Shih-Yang Su, Frank Yu, Michael Zollhรถfer, and Helge Rhodin
Thirty-Fifth Conference on Neural Information Processing Systems (NeurIPS 2021)

Setup

Setup environment

conda create -n anerf python=3.8
conda activate anerf

# install pytorch for your corresponding CUDA environments
pip install torch

# install pytorch3d: note that doing `pip install pytorch3d` directly may install an older version with bugs.
# be sure that you specify the version that matches your CUDA environment. See: https://github.com/facebookresearch/pytorch3d
pip install pytorch3d -f https://dl.fbaipublicfiles.com/pytorch3d/packaging/wheels/py38_cu102_pyt190/download.html

# install other dependencies
pip install -r requirements.txt

Download pre-processed data and pre-trained models

We provide pre-processed data in .h5 format, as well as pre-trained characters for SURREAL and Mixamo dataset.

Please see data/README.md for details.

Testing

You can use run_render.py to render the learned models under different camera motions, or retarget the character to different poses by

python run_render.py --nerf_args logs/surreal_model/args.txt --ckptpath logs/surreal_model/150000.tar \
                     --dataset surreal --entry hard --render_type bullet --render_res 512 512 \
                     --white_bkgd --runname surreal_bullet

Here,

  • --dataset specifies the data source for poses,
  • --entry specifices the particular subset from the dataset to render,
  • --render_type defines the camera motion to use, and
  • --render_res specifies the height and width of the rendered images.

Therefore, the above command will render 512x512 the learned SURREAL character with bullet-time effect like the following (resizsed to 256x256):

The output can be found in render_output/surreal_bullet/.

You can also extract mesh for the learned character:

python run_render.py --nerf_args logs/surreal_model/args.txt --ckptpath logs/surreal_model/150000.tar \
                     --dataset surreal --entry hard --render_type mesh --runname surreal_mesh

You can find the extracted .ply files in render_output/surreal_mesh/meshes/.

To render the mesh as in the paper, run

python render_mesh.py --expname surreal_mesh 

which will output the rendered images in render_output/surreal_mesh/mesh_render/ like the following:

You can change the setting in run_render.py to create your own rendering configuration.

Training

We provide template training configurations in configs/ for different settings.

To train A-NeRF on our pre-processed SURREAL dataset,

python run_nerf.py --config configs/surreal/surreal.txt --basedir logs  --expname surreal_model

The trained weights and log can be found in logs/surreal_model.

To train A-NeRF on our pre-processed Mixamo dataset with estimated poses, run

python run_nerf.py --config configs/mixamo/mixamo.txt --basedir log_mixamo/ --num_workers 8 --subject archer --expname mixamo_archer

This will train A-NeRF on Mixamo Archer with pose refinement for 500k iterations, with 8 worker threads for the dataloader.

You can also add --use_temp_loss --temp_coef 0.05 to optimize the pose with temporal constraint.

Additionally, you can specify --opt_pose_stop 200000 to stop the pose refinement at 200k iteraions to only optimize the body models for the remaining iterations.

To finetune the learned model, run

python run_nerf.py --config configs/mixamo/mixamo_finetune.txt --finetune --ft_path log_mixamo/mixamo_archer/500000.tar --expname mixamo_archer_finetune

This will finetune the learned Mixamo Archer for 200k with the already refined poses. Note that the pose will not be updated during this time.

Citation

@inproceedings{su2021anerf,
    title={A-NeRF: Articulated Neural Radiance Fields for Learning Human Shape, Appearance, and Pose},
    author={Su, Shih-Yang and Yu, Frank and Zollh{\"o}fer, Michael and Rhodin, Helge},
    booktitle = {Advances in Neural Information Processing Systems},
    year={2021}
}

Acknowledgements

Owner
Shih-Yang Su
Enjoy working on ML/RL/CV/MIR related domain.
Shih-Yang Su
Pytorch implementation for "Implicit Semantic Response Alignment for Partial Domain Adaptation"

Implicit-Semantic-Response-Alignment Pytorch implementation for "Implicit Semantic Response Alignment for Partial Domain Adaptation" Prerequisites pyt

4 Dec 19, 2022
This is the repository for Learning to Generate Piano Music With Sustain Pedals

SusPedal-Gen This is the official repository of Learning to Generate Piano Music With Sustain Pedals Demo Page Dataset The dataset used in this projec

Joann Ching 12 Sep 02, 2022
[WACV 2022] Contextual Gradient Scaling for Few-Shot Learning

CxGrad - Official PyTorch Implementation Contextual Gradient Scaling for Few-Shot Learning Sanghyuk Lee, Seunghyun Lee, and Byung Cheol Song In WACV 2

Sanghyuk Lee 4 Dec 05, 2022
An energy estimator for eyeriss-like DNN hardware accelerator

Energy-Estimator-for-Eyeriss-like-Architecture- An energy estimator for eyeriss-like DNN hardware accelerator This is an energy estimator for eyeriss-

HEXIN BAO 2 Mar 26, 2022
Data Augmentation with Variational Autoencoders

Documentation Pyraug This library provides a way to perform Data Augmentation using Variational Autoencoders in a reliable way even in challenging con

112 Nov 30, 2022
A clean implementation based on AlphaZero for any game in any framework + tutorial + Othello/Gobang/TicTacToe/Connect4 and more

Alpha Zero General (any game, any framework!) A simplified, highly flexible, commented and (hopefully) easy to understand implementation of self-play

Surag Nair 3.1k Jan 05, 2023
This repository contains the code for the paper ``Identifiable VAEs via Sparse Decoding''.

Sparse VAE This repository contains the code for the paper ``Identifiable VAEs via Sparse Decoding''. Data Sources The datasets used in this paper wer

Gemma Moran 17 Dec 12, 2022
Semi-Supervised Semantic Segmentation with Pixel-Level Contrastive Learning from a Class-wise Memory Bank

This repository provides the official code for replicating experiments from the paper: Semi-Supervised Semantic Segmentation with Pixel-Level Contrast

Iñigo Alonso Ruiz 58 Dec 15, 2022
The repository forked from NVlabs uses our data. (Differentiable rasterization applied to 3D model simplification tasks)

nvdiffmodeling [origin_code] Differentiable rasterization applied to 3D model simplification tasks, as described in the paper: Appearance-Driven Autom

Qiujie (Jay) Dong 2 Oct 31, 2022
This is the official Pytorch implementation of "Lung Segmentation from Chest X-rays using Variational Data Imputation", Raghavendra Selvan et al. 2020

README This is the official Pytorch implementation of "Lung Segmentation from Chest X-rays using Variational Data Imputation", Raghavendra Selvan et a

Raghav 42 Dec 15, 2022
Volumetric Correspondence Networks for Optical Flow, NeurIPS 2019.

VCN: Volumetric correspondence networks for optical flow [project website] Requirements python 3.6 pytorch 1.1.0-1.3.0 pytorch correlation module (opt

Gengshan Yang 144 Dec 06, 2022
Classification of Long Sequential Data using Circular Dilated Convolutional Neural Networks

Classification of Long Sequential Data using Circular Dilated Convolutional Neural Networks arXiv preprint: https://arxiv.org/abs/2201.02143. Architec

19 Nov 30, 2022
Cycle Consistent Adversarial Domain Adaptation (CyCADA)

Cycle Consistent Adversarial Domain Adaptation (CyCADA) A pytorch implementation of CyCADA. If you use this code in your research please consider citi

Hyunwoo Ko 2 Jan 10, 2022
Flow is a computational framework for deep RL and control experiments for traffic microsimulation.

Flow Flow is a computational framework for deep RL and control experiments for traffic microsimulation. See our website for more information on the ap

867 Jan 02, 2023
SeqFormer: a Frustratingly Simple Model for Video Instance Segmentation

SeqFormer: a Frustratingly Simple Model for Video Instance Segmentation SeqFormer SeqFormer: a Frustratingly Simple Model for Video Instance Segmentat

Junfeng Wu 298 Dec 22, 2022
Official Code Release for "TIP-Adapter: Training-free clIP-Adapter for Better Vision-Language Modeling"

Official Code Release for "TIP-Adapter: Training-free clIP-Adapter for Better Vision-Language Modeling" Pipeline of Tip-Adapter Tip-Adapter can provid

peng gao 187 Dec 28, 2022
CSPML (crystal structure prediction with machine learning-based element substitution)

CSPML (crystal structure prediction with machine learning-based element substitution) CSPML is a unique methodology for the crystal structure predicti

8 Dec 20, 2022
A platform for intelligent agent learning based on a 3D open-world FPS game developed by Inspir.AI.

Wilderness Scavenger: 3D Open-World FPS Game AI Challenge This is a platform for intelligent agent learning based on a 3D open-world FPS game develope

46 Nov 24, 2022
AI-UPV at IberLEF-2021 EXIST task: Sexism Prediction in Spanish and English Tweets Using Monolingual and Multilingual BERT and Ensemble Models

AI-UPV at IberLEF-2021 EXIST task: Sexism Prediction in Spanish and English Tweets Using Monolingual and Multilingual BERT and Ensemble Models Descrip

Angel de Paula 1 Jun 08, 2022
GPOEO is a micro-intrusive GPU online energy optimization framework for iterative applications

GPOEO GPOEO is a micro-intrusive GPU online energy optimization framework for iterative applications. We also implement ODPP [1] as a comparison. [1]

瑞雪轻飏 8 Sep 10, 2022