app.py

from ultralytics import YOLO
from PIL import Image
import gradio as gr
from huggingface_hub import snapshot_download
import os
import cv2
import tempfile


def load_model(repo_id):
    download_dir = snapshot_download(repo_id)
    print(download_dir)
    path = os.path.join(download_dir, "best_int8_openvino_model")
    print(path)
    detection_model = YOLO(path, task='detect')
    return detection_model


def predict_image(pilimg):
    source = pilimg
    # x = np.asarray(pilimg)
    # print(x.shape)
    result = detection_model.predict(source, conf=0.5, iou=0.6)
    img_bgr = result[0].plot()
    out_pilimg = Image.fromarray(img_bgr[..., ::-1])  # RGB-order PIL image
    
    return out_pilimg


def predict_video(video_file):
    cap = cv2.VideoCapture(video_file)
    frame_width = int(cap.get(cv2.CAP_PROP_FRAME_WIDTH))
    frame_height = int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT))
    fps = cap.get(cv2.CAP_PROP_FPS)

    temp_video = tempfile.NamedTemporaryFile(delete=False, suffix=".mp4")
    temp_video_path = temp_video.name
    writer = cv2.VideoWriter(temp_video_path, cv2.VideoWriter_fourcc(*'mp4v'), fps, (frame_width, frame_height))

    while cap.isOpened():
        ret, frame = cap.read()
        if not ret:
            break
        
        results = detection_model.predict(frame, conf=0.5, iou=0.6)
        annotated_frame = results[0].plot()
        writer.write(annotated_frame)

    cap.release()
    writer.release()

    return temp_video_path

REPO_ID = "kks24/highlighter-pen-post-it-notes"
detection_model = load_model(REPO_ID)

with gr.Blocks() as app:
    gr.Markdown("Highlighter Pen & Post-it Notes Detector - 4183940K")
    gr.Markdown("Upload an image or a video to detect Highlighter Pen & Post-it Notes")
    
    with gr.Tabs():
        with gr.Tab("Image Object Detection"):
            gr.Markdown("### Upload an Image to detect Highlighter Pen & Post-it Notes")
            img_input = gr.Image(type="pil", label="Upload Image")
            img_output = gr.Image(type="pil", label="Download Image")
            img_button = gr.Button("Process Image")
            img_button.click(predict_image, inputs=img_input, outputs=img_output)
        
        with gr.Tab("Video Object Detection"):
            gr.Markdown("### Upload a video to detect Highlighter Pen & Post-it Notes. You can download the processed video after detection.")
            video_input = gr.Video(label="Upload Video")
            video_output = gr.File(label="Download Processed Video")  # File download component
            process_button = gr.Button("Process Video")

            process_button.click(predict_video, inputs=video_input, outputs=video_output)

app.launch(share=True)