This repository contains the official implementation for the following paper:
Transformers as Meta-Learners for Implicit Neural Representations
Yinbo Chen, Xiaolong Wang
ECCV 2022
Project page: https://yinboc.github.io/trans-inr/.
@inproceedings{chen2022transinr,
title={Transformers as Meta-Learners for Implicit Neural Representations},
author={Chen, Yinbo and Wang, Xiaolong},
booktitle={European Conference on Computer Vision},
year={2022},
}
- Python 3
- Pytorch 1.12.0
- pyyaml numpy tqdm imageio TensorboardX wandb einops
mkdir data and put different dataset folders in it.
-
CelebA: download (from kaggle), extract, and rename the folder as
celeba(so that images are indata/celeba/img_align_celeba/img_align_celeba). -
Imagenette: download, extract, and rename the folder as
imagenette. -
View synthesis: download from google drive (provided by learnit) and put them in a folder named
learnit_shapenet, unzip the category folders and rename them aschairs,cars,lampscorrespondingly.
Run CUDA_VISIBLE_DEVICES=[GPU] python run_trainer.py --cfg [CONFIG], configs are in cfgs/.
To enable wandb, complete wandb.yaml (in root) and add -w to the training command.
When running multiple multi-gpu training processes, specify -p with different values (0,1,2...) for different ports.
For image reconstruction, test PSNR is automatically evaluated in the training script.
For view synthesis, run in a single GPU with configs in cfgs/nvs_eval. To enable test-time optimization, uncomment (remove #) tto_steps in configs.