lama
1.0.0
Oleh Roman Suvorov, Elizaveta LoGacheva, Anton Mashikhin, Anastasia Remizova, Arsenii Ashukha, Aleksei Silvestrov, Naejin Kong, Harshith Goka, Taman Kiwoong, Victor Lempitsky.
LAMA menggeneralisasi secara mengejutkan dengan baik untuk resolusi yang jauh lebih tinggi (~ 2k❗️) daripada yang dilihat selama pelatihan (256x256), dan mencapai kinerja yang sangat baik bahkan dalam skenario yang menantang, misalnya penyelesaian struktur periodik.
[Halaman Proyek] [Arxiv] [Tambahan] [Bibtex] [Ringkasan Kertas Gan Casual]
Cobalah di Google Colab
(Jangan ragu untuk membagikan makalah Anda dengan membuat masalah)
(Jangan ragu untuk membagikan aplikasi/implementasi/demo Anda dengan membuat masalah)
Kloning Repo: git clone https://github.com/advimman/lama.git
Ada tiga pilihan lingkungan:
Python Virtualenv:
virtualenv inpenv --python=/usr/bin/python3
source inpenv/bin/activate
pip install torch==1.8.0 torchvision==0.9.0
cd lama
pip install -r requirements.txt
Conda
% Install conda for Linux, for other OS download miniconda at https://docs.conda.io/en/latest/miniconda.html
wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh
bash Miniconda3-latest-Linux-x86_64.sh -b -p $HOME/miniconda
$HOME/miniconda/bin/conda init bash
cd lama
conda env create -f conda_env.yml
conda activate lama
conda install pytorch torchvision torchaudio cudatoolkit=10.2 -c pytorch -y
pip install pytorch-lightning==1.2.9
Docker: Tidak diperlukan tindakan ?.
Berlari
cd lama
export TORCH_HOME=$(pwd) && export PYTHONPATH=$(pwd)
1. Unduh model pra-terlatih
Model terbaik (Places2, Places Challenge):
curl -LJO https://huggingface.co/smartywu/big-lama/resolve/main/big-lama.zip
unzip big-lama.zip
Semua model (tempat & celeba-hq):
download [https://drive.google.com/drive/folders/1B2x7eQDgecTL0oh3LSIBDGj0fTxs6Ips?usp=drive_link]
unzip lama-models.zip
2. Siapkan gambar dan topeng
Unduh Test Images:
unzip LaMa_test_images.zip
image1_mask001.png
image1.png
image2_mask001.png
image2.png
image_suffix , mis. .png atau .jpg atau _input.jpg di configs/prediction/default.yaml .3. Prediksi
Di mesin host:
python3 bin/predict.py model.path=$(pwd)/big-lama indir=$(pwd)/LaMa_test_images outdir=$(pwd)/output
Atau di Docker
Perintah berikut akan menarik gambar Docker dari Docker Hub dan menjalankan skrip prediksi
bash docker/2_predict.sh $(pwd)/big-lama $(pwd)/LaMa_test_images $(pwd)/output device=cpu
Docker Cuda:
bash docker/2_predict_with_gpu.sh $(pwd)/big-lama $(pwd)/LaMa_test_images $(pwd)/output
4. Memprediksi dengan penyempurnaan
Di mesin host:
python3 bin/predict.py refine=True model.path=$(pwd)/big-lama indir=$(pwd)/LaMa_test_images outdir=$(pwd)/output
Pastikan Anda menjalankan:
cd lama
export TORCH_HOME=$(pwd) && export PYTHONPATH=$(pwd)
Kemudian unduh model untuk kehilangan perseptual :
mkdir -p ade20k/ade20k-resnet50dilated-ppm_deepsup/
wget -P ade20k/ade20k-resnet50dilated-ppm_deepsup/ http://sceneparsing.csail.mit.edu/model/pytorch/ade20k-resnet50dilated-ppm_deepsup/encoder_epoch_20.pth
Di mesin host:
# Download data from http://places2.csail.mit.edu/download.html
# Places365-Standard: Train(105GB)/Test(19GB)/Val(2.1GB) from High-resolution images section
wget http://data.csail.mit.edu/places/places365/train_large_places365standard.tar
wget http://data.csail.mit.edu/places/places365/val_large.tar
wget http://data.csail.mit.edu/places/places365/test_large.tar
# Unpack train/test/val data and create .yaml config for it
bash fetch_data/places_standard_train_prepare.sh
bash fetch_data/places_standard_test_val_prepare.sh
# Sample images for test and viz at the end of epoch
bash fetch_data/places_standard_test_val_sample.sh
bash fetch_data/places_standard_test_val_gen_masks.sh
# Run training
python3 bin/train.py -cn lama-fourier location=places_standard
# To evaluate trained model and report metrics as in our paper
# we need to sample previously unseen 30k images and generate masks for them
bash fetch_data/places_standard_evaluation_prepare_data.sh
# Infer model on thick/thin/medium masks in 256 and 512 and run evaluation
# like this:
python3 bin/predict.py
model.path=$(pwd)/experiments/<user>_<date:time>_lama-fourier_/
indir=$(pwd)/places_standard_dataset/evaluation/random_thick_512/
outdir=$(pwd)/inference/random_thick_512 model.checkpoint=last.ckpt
python3 bin/evaluate_predicts.py
$(pwd)/configs/eval2_gpu.yaml
$(pwd)/places_standard_dataset/evaluation/random_thick_512/
$(pwd)/inference/random_thick_512
$(pwd)/inference/random_thick_512_metrics.csv
Docker: Todo
Di mesin host:
# Make shure you are in lama folder
cd lama
export TORCH_HOME=$(pwd) && export PYTHONPATH=$(pwd)
# Download CelebA-HQ dataset
# Download data256x256.zip from https://drive.google.com/drive/folders/11Vz0fqHS2rXDb5pprgTjpD7S2BAJhi1P
# unzip & split into train/test/visualization & create config for it
bash fetch_data/celebahq_dataset_prepare.sh
# generate masks for test and visual_test at the end of epoch
bash fetch_data/celebahq_gen_masks.sh
# Run training
python3 bin/train.py -cn lama-fourier-celeba data.batch_size=10
# Infer model on thick/thin/medium masks in 256 and run evaluation
# like this:
python3 bin/predict.py
model.path=$(pwd)/experiments/<user>_<date:time>_lama-fourier-celeba_/
indir=$(pwd)/celeba-hq-dataset/visual_test_256/random_thick_256/
outdir=$(pwd)/inference/celeba_random_thick_256 model.checkpoint=last.ckpt
Docker: Todo
Di mesin host:
# This script downloads multiple .tar files in parallel and unpacks them
# Places365-Challenge: Train(476GB) from High-resolution images (to train Big-Lama)
bash places_challenge_train_download.sh
TODO: prepare
TODO: train
TODO: eval
Docker: Todo
Silakan periksa skrip bash untuk persiapan data dan pembuatan topeng dari bagian Celebahq, jika Anda terjebak di salah satu langkah berikut.
Di mesin host:
# Make shure you are in lama folder
cd lama
export TORCH_HOME=$(pwd) && export PYTHONPATH=$(pwd)
# You need to prepare following image folders:
$ ls my_dataset
train
val_source # 2000 or more images
visual_test_source # 100 or more images
eval_source # 2000 or more images
# LaMa generates random masks for the train data on the flight,
# but needs fixed masks for test and visual_test for consistency of evaluation.
# Suppose, we want to evaluate and pick best models
# on 512x512 val dataset with thick/thin/medium masks
# And your images have .jpg extention:
python3 bin/gen_mask_dataset.py
$(pwd)/configs/data_gen/random_<size>_512.yaml # thick, thin, medium
my_dataset/val_source/
my_dataset/val/random_<size>_512.yaml # thick, thin, medium
--ext jpg
# So the mask generator will:
# 1. resize and crop val images and save them as .png
# 2. generate masks
ls my_dataset/val/random_medium_512/
image1_crop000_mask000.png
image1_crop000.png
image2_crop000_mask000.png
image2_crop000.png
...
# Generate thick, thin, medium masks for visual_test folder:
python3 bin/gen_mask_dataset.py
$(pwd)/configs/data_gen/random_<size>_512.yaml #thick, thin, medium
my_dataset/visual_test_source/
my_dataset/visual_test/random_<size>_512/ #thick, thin, medium
--ext jpg
ls my_dataset/visual_test/random_thick_512/
image1_crop000_mask000.png
image1_crop000.png
image2_crop000_mask000.png
image2_crop000.png
...
# Same process for eval_source image folder:
python3 bin/gen_mask_dataset.py
$(pwd)/configs/data_gen/random_<size>_512.yaml #thick, thin, medium
my_dataset/eval_source/
my_dataset/eval/random_<size>_512/ #thick, thin, medium
--ext jpg
# Generate location config file which locate these folders:
touch my_dataset.yaml
echo "data_root_dir: $(pwd)/my_dataset/" >> my_dataset.yaml
echo "out_root_dir: $(pwd)/experiments/" >> my_dataset.yaml
echo "tb_dir: $(pwd)/tb_logs/" >> my_dataset.yaml
mv my_dataset.yaml ${PWD}/configs/training/location/
# Check data config for consistency with my_dataset folder structure:
$ cat ${PWD}/configs/training/data/abl-04-256-mh-dist
...
train:
indir: ${location.data_root_dir}/train
...
val:
indir: ${location.data_root_dir}/val
img_suffix: .png
visual_test:
indir: ${location.data_root_dir}/visual_test
img_suffix: .png
# Run training
python3 bin/train.py -cn lama-fourier location=my_dataset data.batch_size=10
# Evaluation: LaMa training procedure picks best few models according to
# scores on my_dataset/val/
# To evaluate one of your best models (i.e. at epoch=32)
# on previously unseen my_dataset/eval do the following
# for thin, thick and medium:
# infer:
python3 bin/predict.py
model.path=$(pwd)/experiments/<user>_<date:time>_lama-fourier_/
indir=$(pwd)/my_dataset/eval/random_<size>_512/
outdir=$(pwd)/inference/my_dataset/random_<size>_512
model.checkpoint=epoch32.ckpt
# metrics calculation:
python3 bin/evaluate_predicts.py
$(pwd)/configs/eval2_gpu.yaml
$(pwd)/my_dataset/eval/random_<size>_512/
$(pwd)/inference/my_dataset/random_<size>_512
$(pwd)/inference/my_dataset/random_<size>_512_metrics.csv
Atau di Docker:
TODO: train
TODO: eval
Perintah berikut akan menjalankan skrip yang menghasilkan topeng acak.
bash docker/1_generate_masks_from_raw_images.sh
configs/data_gen/random_medium_512.yaml
/directory_with_input_images
/directory_where_to_store_images_and_masks
--ext png
Perintah pembuatan data uji menyimpan gambar dalam format, yang cocok untuk prediksi.
Tabel di bawah ini menjelaskan konfigurasi mana yang kami gunakan untuk menghasilkan set uji yang berbeda dari kertas. Perhatikan bahwa kami tidak memperbaiki benih acak , sehingga hasilnya akan sedikit berbeda setiap kali.
| Tempat 512x512 | Celeba 256x256 | |
|---|---|---|
| Sempit | acak_thin_512.yaml | acak_thin_256.yaml |
| Sedang | acak_medium_512.yaml | Random_Medium_256.yaml |
| Lebar | acak_thick_512.yaml | acak_thick_256.yaml |
Jangan ragu untuk mengubah jalur konfigurasi (argumen #1) ke konfigurasi lain di configs/data_gen atau menyesuaikan file konfigurasi itu sendiri.
Anda juga dapat mengganti parameter dalam konfigurasi seperti ini:
python3 bin/train.py -cn <config> data.batch_size=10 run_title=my-title
Di mana ekstensi file .yaml dihilangkan
Nama konfigurasi untuk model dari kertas (substititual ke dalam perintah pelatihan):
* big-lama
* big-lama-regular
* lama-fourier
* lama-regular
* lama_small_train_masks
Yang duduk di konfigurasi/pelatihan/folder
Todo
Jika Anda menemukan kode ini bermanfaat, harap pertimbangkan mengutip:
@article{suvorov2021resolution,
title={Resolution-robust Large Mask Inpainting with Fourier Convolutions},
author={Suvorov, Roman and Logacheva, Elizaveta and Mashikhin, Anton and Remizova, Anastasia and Ashukha, Arsenii and Silvestrov, Aleksei and Kong, Naejin and Goka, Harshith and Park, Kiwoong and Lempitsky, Victor},
journal={arXiv preprint arXiv:2109.07161},
year={2021}
}
