app.py

import argparse
import numpy as np
import cv2
import os
import gradio as gr
import insightface
from insightface.app import FaceAnalysis
from dofaker import FaceSwapper, PoseSwapper

def parse_args():
    parser = argparse.ArgumentParser(description='⏳ Running face swap')
    parser.add_argument(
        '--inbrowser',
        help='Automatically launch the interface in a new tab on the default browser.',
        dest='inbrowser',
        default=True)
    parser.add_argument(
        '--server_port',
        help='Start gradio app on this port (if available).',
        dest='server_port',
        type=int,
        default=None)
    return parser.parse_args()

def swap_face(input_path, dst_path, src_path, use_enhancer, use_sr, scale, face_sim_thre):
    faker = FaceSwapper(use_enhancer=use_enhancer, use_sr=use_sr, scale=scale, face_sim_thre=face_sim_thre)
    output_path = faker.run(input_path, dst_path, src_path)
    return output_path

def swap_pose(input_path, target_path, use_enhancer, use_sr, scale):
    faker = PoseSwapper(use_enhancer=use_enhancer, use_sr=use_sr, scale=scale)
    output_path = faker.run(input_path, target_path)
    return output_path

def add_bbox_padding(bbox, margin=5):
    return [
        bbox[0] - margin,
        bbox[1] - margin,
        bbox[2] + margin,
        bbox[3] + margin
    ]

def select_handler(img, evt):
    faces = app.get(img)
    faces = sorted(faces, key=lambda x: x.bbox[0])
    cropped_image = []
    face_index = -1
    sel_face_index = 0
    for face in faces:
        box = face.bbox.astype(np.int32)
        face_index += 1
        if point_in_box((box[0], box[1]), (box[2], box[3]), (evt.index[0], evt.index[1])):
            margin = int((box[2] - box[0]) * 0.35)
            box = add_bbox_padding(box, margin)
            box = np.clip(box, 0, None)
            sel_face_index = face_index            
            cropped_image = img[box[1]:box[3], box[0]:box[2]]
    return cropped_image, sel_face_index

def point_in_box(bl, tr, p):
    return bl[0] < p[0] < tr[0] and bl[1] < p[1] < tr[1]

def get_faces(img):
    faces = app.get(img)
    faces = sorted(faces, key=lambda x: x.bbox[0])
    return img, len(faces)

def main():
    args = parse_args()
    app = FaceAnalysis(name='buffalo_l')
    app.prepare(ctx_id=0, det_size=(640, 640))
    swapper = insightface.model_zoo.get_model('inswapper_128.onnx', download=True, download_zip=True)

    with gr.Blocks(theme='ysharma/huggingface@=0.0.4') as web_ui:
        gr.Markdown('# 👑 FaceClone')                              
        gr.Markdown('### Deepfake Swap Face and Pose.')
        
        with gr.Tab('🧩 FaceSwapper'):
            gr.Markdown(' 🧸 FaceClone: Powered by Gradio')
            with gr.Tab('📥 Face'):
                with gr.Row():
                    with gr.Column():
                        gr.Markdown('🖇️ Source Face to be swapped')
                        image_input = gr.Image(type='filepath')
                        with gr.Row():
                            with gr.Column():
                                gr.Markdown('🎭 Target source included with Face')
                                dst_face_image = gr.Image(type='filepath')
                            with gr.Column():
                                gr.Markdown('🧑‍🦳 Source Face to Replace Target Face')
                                src_face_image = gr.Image(type='filepath')
                    with gr.Column():
                        output_image = gr.Image(type='filepath')
                        use_enhancer = gr.Checkbox(label="👑 Face Enhance", info="🧿 Whether use face enhance model.")
                        with gr.Row():
                            use_sr = gr.Checkbox(label="🛠️ Super resolution", info="⚒️ Whether use image resolution model.")
                            scale = gr.Number(value=1, label='📐 Image super resolution scale')
                        with gr.Row():
                            face_sim_thre = gr.Number(value=0.6, label='🧲 Face similarity threshold', minimum=0.0, maximum=1.0)
                        convert_button = gr.Button('🔍 Swap')
                        convert_button.click(fn=swap_face, inputs=[image_input, dst_face_image, src_face_image, use_enhancer, use_sr, scale, face_sim_thre], outputs=[output_image])

            with gr.Tab('🖥️ Video'):
                with gr.Row():
                    with gr.Column():
                        gr.Markdown('📺 Target Video')
                        video_input = gr.Video()
                        with gr.Row():
                            with gr.Column():
                                gr.Markdown('👳 Target Face')
                                dst_face_image = gr.Image(type='filepath')
                            with gr.Column():
                                gr.Markdown('🫅 Source Face')
                                src_face_image = gr.Image(type='filepath')
                    with gr.Column():
                        output_video = gr.Video()
                        use_enhancer = gr.Checkbox(label="📸 Face Enhance", info="📷 Whether use face enhance model.")
                        with gr.Row():
                            use_sr = gr.Checkbox(label="🔋 Super Resolution", info="🔦 Whether use image resolution model.")
                            scale = gr.Number(value=1, label='🗞️ Super Resolution Image')
                        with gr.Row():
                            face_sim_thre = gr.Number(value=0.6, label='💌 Face similarity threshold', minimum=0.0, maximum=1.0)
                        convert_button = gr.Button('🎯 Swap')
                        convert_button.click(fn=swap_face, inputs=[video_input, dst_face_image, src_face_image, use_enhancer, use_sr, scale, face_sim_thre], outputs=[output_video])

        with gr.Tab('🪩 PoseSwapper'):
            gr.Markdown(' 🧸 FaceClone: Deepfake powered by Gradio')
            with gr.Tab('📷 Image'):
                with gr.Row():
                    with gr.Column():
                        gr.Markdown('🧑‍🦳 The source image to be swapped')
                        image_input = gr.Image(type='filepath')
                        gr.Markdown('🗳️ The target image with pose')
                        target = gr.Image(type='filepath')
                    with gr.Column():
                        output_image = gr.Image(type='filepath')
                        use_enhancer = gr.Checkbox(label="👑 Face Enhance", info="🧾 Whether use face enhance model.")
                        with gr.Row():
                            use_sr = gr.Checkbox(label="🛠️ Super Resolution", info="⌛ Whether use image resolution model.")
                            scale = gr.Number(value=1, label='📸 Super resolution scale')
                        convert_button = gr.Button('🎯 Swap')
                        convert_button.click(fn=swap_pose, inputs=[image_input, target, use_enhancer, use_sr, scale], outputs=[output_image])

    web_ui.launch(inbrowser=args.inbrowser, server_port=args.server_port)

if __name__ == '__main__':
    main()