app.py

from utils.configuration import Configuration
import tensorflow as tf
from utils.model import ModelLoss
from utils.model import LFUNet
from utils.architectures import UNet

import gradio as gr

configuration = Configuration()
filters = (64, 128, 128, 256, 256, 512)
kernels = (7, 7, 7, 3, 3, 3)
input_image_size = (256, 256, 3)
architecture = UNet.RESIDUAL_ATTENTION_UNET_SEPARABLE_CONV

trained_model = LFUNet.build_model(architecture=architecture, input_size=input_image_size, filters=filters,
                                   kernels=kernels, configuration=configuration)
trained_model.compile(
            loss=ModelLoss.ms_ssim_l1_perceptual_loss,
            optimizer=tf.keras.optimizers.Adam(1e-4),
            metrics=["acc", tf.keras.metrics.Recall(), tf.keras.metrics.Precision()]
        )

weights_path = "model_weights/model_epochs-40_batch-20_loss-ms_ssim_l1_perceptual_loss_20230210_15_45_38.ckpt"
trained_model.load_weights(weights_path)

def main(input_img):
    try:
        print(input_img)
        predicted_image = trained_model.predict(input_img)
        return predicted_image
    except Exception as e:
        raise gr.Error("Sorry, something went wrong. Please try again!")

demo = gr.Interface(
    title= "Lightweight network for face unmasking",
    description= "This is a demo of a <b>Lightweight network for face unmasking</b> \
        designed to provide a powerful and efficient solution for restoring facial details obscured by masks.<br> \
        To use it, simply upload your image, or click one of the examples to load them. Inference needs some time since this demo uses CPU.",
    fn = main, 
    inputs= gr.Image(type="filepath").style(height=256),
    outputs=gr.Image(type='numpy',shape=(256, 256, 3)).style(height=256),
    # allow_flagging='never',
    examples=[
        ["examples/1.png"],
        ["examples/2.png"],
        ["examples/3.png"],
        ["examples/4.png"],
        ["examples/5.png"],
        ["examples/6.png"],
        ["examples/7.png"],
        ["examples/8.png"],
    ],
    css = """
        .svelte-mppz8v {
            text-align: -webkit-center;
        }

        .gallery {
            display: flex;
            flex-wrap: wrap;
            width: 100%;
        }

        p {
            font-size: medium;
        }

        h1 {
            font-size: xx-large;
        }
        """,
    theme= 'EveryPizza/Cartoony-Gradio-Theme',
    # article = "<p style='text-align: center'><a href='https://arxiv.org/abs/2204.04676' target='_blank'>Simple Baselines for Image Restoration</a> | <a href='https://arxiv.org/abs/2204.08714' target='_blank'>NAFSSR: Stereo Image Super-Resolution Using NAFNet</a>  | <a href='https://github.com/megvii-research/NAFNet' target='_blank'> Github Repo</a></p>"
)
demo.launch(show_error=True, share= True)