app.py

#!/usr/bin/env python3
import json
import sys
import os
import io
import argparse
import uuid
import base64
import logging
import time
import copy
import cv2
import insightface
import numpy as np
from typing import List, Union
from PIL import Image
from restoration import *
from flask import Flask, request, jsonify, make_response
from waitress import serve

LOG_LEVEL = logging.DEBUG
TMP_PATH = '/tmp/inswapper'
script_dir = os.path.dirname(os.path.abspath(__file__))
log_path = ''

# Mac does not have permission to /var/log for example
if sys.platform == 'linux':
    log_path = '/var/log/'

logging.basicConfig(
    filename=f'{log_path}inswapper.log',
    format='%(asctime)s : %(levelname)s : %(message)s',
    level=LOG_LEVEL
)

logging.getLogger().addHandler(logging.StreamHandler(sys.stdout))


def process_request(request_obj):
    try:
        logging.debug('Swapping face')
        face_swap_timer = Timer()
        result_image = face_swap(request_obj['source_image'], request_obj['target_image'])
        face_swap_time = face_swap_timer.get_elapsed_time()
        logging.info(f'Time taken to swap face: {face_swap_time} seconds')

        response = {
            'status': 'ok',
            'image': result_image
        }
    except Exception as e:
        logging.error(e)
        response = {
            'status': 'error',
            'msg': 'Face swap failed',
            'detail': str(e)
        }

    return response


class Timer:
    def __init__(self):
        self.start = time.time()

    def restart(self):
        self.start = time.time()

    def get_elapsed_time(self):
        end = time.time()
        return round(end - self.start, 1)


def get_args():
    parser = argparse.ArgumentParser(
        description='Inswapper REST API'
    )

    parser.add_argument(
        '-p', '--port',
        help='Port to listen on',
        type=int,
        default=80
    )

    parser.add_argument(
        '-H', '--host',
        help='Host to bind to',
        default='0.0.0.0'
    )

    return parser.parse_args()


def determine_file_extension(image_data):
    try:
        if image_data.startswith('/9j/'):
            image_extension = '.jpg'
        elif image_data.startswith('iVBORw0Kg'):
            image_extension = '.png'
        else:
            # Default to png if we can't figure out the extension
            image_extension = '.png'
    except Exception as e:
        image_extension = '.png'

    return image_extension


def write_base64_to_disk(file_b64: str, file_path: str):
    with open(file_path, 'wb') as file:
        file.write(base64.b64decode(file_b64))


def get_face_swap_model(model_path: str):
    model = insightface.model_zoo.get_model(model_path)
    return model


def get_face_analyser(model_path: str,
                      det_size=(320, 320)):
    face_analyser = insightface.app.FaceAnalysis(name="buffalo_l", root="./checkpoints")
    face_analyser.prepare(ctx_id=0, det_size=det_size)
    return face_analyser


def get_one_face(face_analyser,
                 frame:np.ndarray):
    face = face_analyser.get(frame)
    try:
        return min(face, key=lambda x: x.bbox[0])
    except ValueError:
        return None


def get_many_faces(face_analyser,
                   frame:np.ndarray):
    """
    get faces from left to right by order
    """
    try:
        face = face_analyser.get(frame)
        return sorted(face, key=lambda x: x.bbox[0])
    except IndexError:
        return None


def swap_face(face_swapper,
              source_faces,
              target_faces,
              source_index,
              target_index,
              temp_frame):
    """
    paste source_face on target image
    """
    source_face = source_faces[source_index]
    target_face = target_faces[target_index]

    return face_swapper.get(temp_frame, target_face, source_face, paste_back=True)


def process(source_img: Union[Image.Image, List],
            target_img: Image.Image,
            source_indexes: str,
            target_indexes: str,
            model: str):

    # load face_analyser
    face_analyser = get_face_analyser(model)

    # load face_swapper
    model_path = os.path.join(os.path.abspath(os.path.dirname(__file__)), model)
    face_swapper = get_face_swap_model(model_path)

    # read target image
    target_img = cv2.cvtColor(np.array(target_img), cv2.COLOR_RGB2BGR)

    # detect faces that will be replaced in target_img
    target_faces = get_many_faces(face_analyser, target_img)
    num_target_faces = len(target_faces)
    num_source_images = len(source_img)

    if target_faces is not None:
        temp_frame = copy.deepcopy(target_img)
        if isinstance(source_img, list) and num_source_images == num_target_faces:
            logging.debug('Replacing the faces in the target image from left to right by order')
            for i in range(num_target_faces):
                source_faces = get_many_faces(face_analyser, cv2.cvtColor(np.array(source_img[i]), cv2.COLOR_RGB2BGR))
                source_index = i
                target_index = i

                if source_faces is None:
                    raise Exception('No source faces found!')

                temp_frame = swap_face(
                    face_swapper,
                    source_faces,
                    target_faces,
                    source_index,
                    target_index,
                    temp_frame
                )
        elif num_source_images == 1:
            # detect source faces that will be replaced into the target image
            source_faces = get_many_faces(face_analyser, cv2.cvtColor(np.array(source_img[0]), cv2.COLOR_RGB2BGR))
            num_source_faces = len(source_faces)
            logging.debug(f'Source faces: {num_source_faces}')
            logging.debug(f'Target faces: {num_target_faces}')

            if source_faces is None:
                raise Exception('No source faces found!')

            if target_indexes == "-1":
                if num_source_faces == 1:
                    logging.debug('Replacing all faces in target image with the same face from the source image')
                    num_iterations = num_target_faces
                elif num_source_faces < num_target_faces:
                    logging.debug('There are less faces in the source image than the target image, replacing as many as we can')
                    num_iterations = num_source_faces
                elif num_target_faces < num_source_faces:
                    logging.debug('There are less faces in the target image than the source image, replacing as many as we can')
                    num_iterations = num_target_faces
                else:
                    logging.debug('Replacing all faces in the target image with the faces from the source image')
                    num_iterations = num_target_faces

                for i in range(num_iterations):
                    source_index = 0 if num_source_faces == 1 else i
                    target_index = i

                    temp_frame = swap_face(
                        face_swapper,
                        source_faces,
                        target_faces,
                        source_index,
                        target_index,
                        temp_frame
                    )
            elif source_indexes == '-1' and target_indexes == '-1':
                logging.debug('Replacing specific face(s) in the target image with the face from the source image')
                target_indexes = target_indexes.split(',')
                source_index = 0

                for target_index in target_indexes:
                    target_index = int(target_index)

                    temp_frame = swap_face(
                        face_swapper,
                        source_faces,
                        target_faces,
                        source_index,
                        target_index,
                        temp_frame
                    )
            else:
                logging.debug('Replacing specific face(s) in the target image with specific face(s) from the source image')

                if source_indexes == "-1":
                    source_indexes = ','.join(map(lambda x: str(x), range(num_source_faces)))

                if target_indexes == "-1":
                    target_indexes = ','.join(map(lambda x: str(x), range(num_target_faces)))

                source_indexes = source_indexes.split(',')
                target_indexes = target_indexes.split(',')
                num_source_faces_to_swap = len(source_indexes)
                num_target_faces_to_swap = len(target_indexes)

                if num_source_faces_to_swap > num_source_faces:
                    raise Exception('Number of source indexes is greater than the number of faces in the source image')

                if num_target_faces_to_swap > num_target_faces:
                    raise Exception('Number of target indexes is greater than the number of faces in the target image')

                if num_source_faces_to_swap > num_target_faces_to_swap:
                    num_iterations = num_source_faces_to_swap
                else:
                    num_iterations = num_target_faces_to_swap

                if num_source_faces_to_swap == num_target_faces_to_swap:
                    for index in range(num_iterations):
                        source_index = int(source_indexes[index])
                        target_index = int(target_indexes[index])

                        if source_index > num_source_faces-1:
                            raise ValueError(f'Source index {source_index} is higher than the number of faces in the source image')

                        if target_index > num_target_faces-1:
                            raise ValueError(f'Target index {target_index} is higher than the number of faces in the target image')

                        temp_frame = swap_face(
                            face_swapper,
                            source_faces,
                            target_faces,
                            source_index,
                            target_index,
                            temp_frame
                        )
        else:
            logging.error('Unsupported face configuration')
            raise Exception('Unsupported face configuration')
        result = temp_frame
    else:
        logging.error('No target faces found')
        raise Exception('No target faces found!')

    result_image = Image.fromarray(cv2.cvtColor(result, cv2.COLOR_BGR2RGB))
    return result_image


def face_swap(src_img_path,
              target_img_path,
              source_indexes,
              target_indexes,
              background_enhance,
              face_restore,
              face_upsample,
              upscale,
              codeformer_fidelity,
              output_format):

    source_img_paths = src_img_path.split(';')
    source_img = [Image.open(img_path) for img_path in source_img_paths]
    target_img = Image.open(target_img_path)

    # download from https://huggingface.co/ashleykleynhans/inswapper/tree/main
    model = os.path.join(script_dir, 'checkpoints/inswapper_128.onnx')
    logging.debug(f'Face swap model: {model}')

    try:
        logging.debug('Performing face swap')
        result_image = process(
            source_img,
            target_img,
            source_indexes,
            target_indexes,
            model
        )
        logging.debug('Face swap complete')
    except Exception as e:
        raise

    # make sure the ckpts downloaded successfully
    check_ckpts()

    if face_restore:
        # https://huggingface.co/spaces/sczhou/CodeFormer
        logging.debug('Setting upsampler to RealESRGAN_x2plus')
        upsampler = set_realesrgan()

        if torch.cuda.is_available():
            torch_device = 'cuda'
        else:
            torch_device = 'cpu'

        logging.debug(f'Torch device: {torch_device.upper()}')
        device = torch.device(torch_device)

        codeformer_net = ARCH_REGISTRY.get('CodeFormer')(
            dim_embd=512,
            codebook_size=1024,
            n_head=8,
            n_layers=9,
            connect_list=['32', '64', '128', '256'],
        ).to(device)

        ckpt_path = os.path.join(script_dir, 'CodeFormer/CodeFormer/weights/CodeFormer/codeformer.pth')
        logging.debug(f'Loading CodeFormer model: {ckpt_path}')
        checkpoint = torch.load(ckpt_path)['params_ema']
        codeformer_net.load_state_dict(checkpoint)
        codeformer_net.eval()
        result_image = cv2.cvtColor(np.array(result_image), cv2.COLOR_RGB2BGR)
        logging.debug('Performing face restoration using CodeFormer')

        try:
            result_image = face_restoration(
                result_image,
                background_enhance,
                face_upsample,
                upscale,
                codeformer_fidelity,
                upsampler,
                codeformer_net,
                device
            )
        except Exception as e:
            raise

        logging.debug('CodeFormer face restoration completed successfully')
        result_image = Image.fromarray(result_image)

    output_buffer = io.BytesIO()
    result_image.save(output_buffer, format=output_format)
    image_data = output_buffer.getvalue()

    return base64.b64encode(image_data).decode('utf-8')


app = Flask(__name__)


@app.errorhandler(400)
def not_found(error):
    return make_response(jsonify(
        {
            'status': 'error',
            'msg': f'Bad Request',
            'detail': str(error)
        }
    ), 400)


@app.errorhandler(404)
def not_found(error):
    return make_response(jsonify(
        {
            'status': 'error',
            'msg': f'{request.url} not found',
            'detail': str(error)
        }
    ), 404)


@app.errorhandler(500)
def internal_server_error(error):
    return make_response(jsonify(
        {
            'status': 'error',
            'msg': 'Internal Server Error',
            'detail': str(error)
        }
    ), 500)


@app.route('/', methods=['GET'])
def ping():
    return make_response(jsonify(
        {
            'status': 'ok'
        }
    ), 200)


@app.route('/faceswap', methods=['POST'])
def face_swap_api():
    total_timer = Timer()
    logging.debug('Received face swap API request')
    payload = request.get_json()

    if not os.path.exists(TMP_PATH):
        logging.debug(f'Creating temporary directory: {TMP_PATH}')
        os.makedirs(TMP_PATH)

    unique_id = uuid.uuid4()
    source_image_data = payload['source_image']
    target_image_data = payload['target_image']

    # Decode the source image data
    source_image = base64.b64decode(source_image_data)
    source_file_extension = determine_file_extension(source_image_data)
    source_image_path = f'{TMP_PATH}/source_{unique_id}{source_file_extension}'

    # Save the source image to disk
    with open(source_image_path, 'wb') as source_file:
        source_file.write(source_image)

    # Decode the target image data
    target_image = base64.b64decode(target_image_data)
    target_file_extension = determine_file_extension(target_image_data)
    target_image_path = f'{TMP_PATH}/target_{unique_id}{target_file_extension}'

    # Save the target image to disk
    with open(target_image_path, 'wb') as target_file:
        target_file.write(target_image)

    # Set defaults if they are not specified in the payload
    if 'source_indexes' not in payload:
        payload['source_indexes'] = '-1'

    if 'target_indexes' not in payload:
        payload['target_indexes'] = '-1'

    if 'background_enhance' not in payload:
        payload['background_enhance'] = True

    if 'face_restore' not in payload:
        payload['face_restore'] = True

    if 'face_upsample' not in payload:
        payload['face_upsample'] = True

    if 'upscale' not in payload:
        payload['upscale'] = 1

    if 'codeformer_fidelity' not in payload:
        payload['codeformer_fidelity'] = 0.5

    if 'output_format' not in payload:
        payload['output_format'] = 'JPEG'

    try:
        logging.debug(f'Source indexes: {payload["source_indexes"]}')
        logging.debug(f'Target indexes: {payload["target_indexes"]}')
        logging.debug(f'Background enhance: {payload["background_enhance"]}')
        logging.debug(f'Face Restoration: {payload["face_restore"]}')
        logging.debug(f'Face Upsampling: {payload["face_upsample"]}')
        logging.debug(f'Upscale: {payload["upscale"]}')
        logging.debug(f'Codeformer Fidelity: {payload["codeformer_fidelity"]}')
        logging.debug(f'Output Format: {payload["output_format"]}')

        result_image = face_swap(
            source_image_path,
            target_image_path,
            payload['source_indexes'],
            payload['target_indexes'],
            payload['background_enhance'],
            payload['face_restore'],
            payload['face_upsample'],
            payload['upscale'],
            payload['codeformer_fidelity'],
            payload['output_format']
        )

        status_code = 200

        response = {
            'status': 'ok',
            'image': result_image
        }
    except Exception as e:
        logging.error(e)

        response = {
            'status': 'error',
            'msg': 'Face swap failed',
            'detail': str(e)
        }

        status_code = 500

    # Clean up temporary images
    os.remove(source_image_path)
    os.remove(target_image_path)

    total_time = total_timer.get_elapsed_time()
    logging.info(f'Total time taken for face swap API call {total_time} seconds')

    return make_response(jsonify(response), status_code)


if __name__ == '__main__':
    args = get_args()

    serve(
        app,
        host=args.host,
        port=args.port
    )