liuzhen0212/FINER

FINER: Flexible spectral-bias tuning in Implicit NEural Representation by Variable-periodic Activation Functions

FINER: Flexible spectral-bias tuning in Implicit NEural Representation by Variable-periodic Activation Functions

CVPR 2024

Zhen Liu^1,*, Hao Zhu^1,*, Qi Zhang², Jingde Fu¹, Weibing Deng¹, Zhan Ma¹, Yanwen Guo¹, Xun Cao¹,

¹Nanjing University, ²Tencent AI Lab, ^*Equal contibution

Project Page | Paper

We propose a novel implicit neural representation with flexible spectral-bias tuning for representing and optimizing signals. The repo contains the codes for image fitting. For the SDF and NeRF experiments, we utilized the codes of Bacon and torch-ngp, respectively.

Setup

conda create -n finer python=3.8
conda activate finer
pip install -r requirements.txt

Training

Image Fitting

bash run_finer.sh 
# run_siren.sh; run_pemlp.sh; run_gauss.sh; run_wire.sh

Visualizations

• The outputs of the first layer neurons

Citation

@inproceedings{liu2024finer,
    title = {FINER: Flexible spectral-bias tuning in Implicit NEural Representation by Variable-periodic Activation Functions},
    author = {Liu, Zhen and Zhu, Hao and Zhang, Qi and Fu, Jingde and Deng, Weibing and Ma, Zhan and Guo, Yanwen and Cao, Xun},
    booktitle = {Proceedings of the IEEE/CVF Computer Vision and Pattern Recognition Conference (CVPR)},
    year = {2024}
}