Nerdy Rodent's video: Bring Old Photos Back To Life

Bring Old Photos Back to Life from Microsoft Reasearch. Use the power of generative adversarial networks to "bring life" back to images! Removes scratches, dust, and other noise from an image. Includes face detection too! Github: https://github.com/microsoft/Bringing-Old-Photos-Back-to-Life.git If you don’t have an Nvidia gpu, google colab is the answer: https://colab.research.google.com/drive/1NEm6AsybIiC5TwTU_4DqDkQO0nFRB-uA?usp=sharing Anaconda: https://www.anaconda.com/products/individual == Install == git clone https://github.com/microsoft/Bringing-Old-Photos-Back-to-Life.git cd ../Bringing-Old-Photos-Back-to-Life/ conda create --name photolife python=3.7 conda activate photolife cd Face_Enhancement/models/networks/ git clone https://github.com/vacancy/Synchronized-BatchNorm-PyTorch cp -rf Synchronized-BatchNorm-PyTorch/sync_batchnorm . cd ../../../ cd Global/detection_models git clone https://github.com/vacancy/Synchronized-BatchNorm-PyTorch cp -rf Synchronized-BatchNorm-PyTorch/sync_batchnorm . cd ../../ cd Face_Detection/ wget http://dlib.net/files/shape_predictor_68_face_landmarks.dat.bz2 bzip2 -d shape_predictor_68_face_landmarks.dat.bz2 cd ../ cd Face_Enhancement/ wget https://facevc.blob.core.windows.net/zhanbo/old_photo/pretrain/Face_Enhancement/checkpoints.zip unzip checkpoints.zip cd ../ cd Global/ wget https://facevc.blob.core.windows.net/zhanbo/old_photo/pretrain/Global/checkpoints.zip unzip checkpoints.zip cd ../ pip install -r requirements.txt * Nvidia RTX 3000 Series (e.g. RTX 3080) requires CUDA 11, so: conda install pytorch torchvision torchaudio cudatoolkit=11.0 -c pytorch == Run == Full pipeline - Images ~512x512 are OK python run.py --input_folder test_images/old/ --output_folder Output --GPU 0 With scratch requires much more VRAM: python run.py --input_folder test_images/old_w_scratch/ --output_folder Output_ws --GPU 0 --with_scratch