#目标检测
import io
import gradio as gr
import matplotlib.pyplot as plt
import torch
from PIL import Image
from transformers import AutoFeatureExtractor, DetrForObjectDetection
from gradio.themes.utils import sizes


theme = gr.themes.Default(radius_size=sizes.radius_none).set(
    block_label_text_color = '#4D63FF',
    block_title_text_color = '#4D63FF',
    button_primary_text_color = '#4D63FF',
    button_primary_background_fill='#FFFFFF',
    button_primary_border_color='#4D63FF',
    button_primary_background_fill_hover='#EDEFFF',
)


def make_prediction(img, feature_extractor, model):
    inputs = feature_extractor(img, return_tensors="pt")
    outputs = model(**inputs)
    img_size = torch.tensor([tuple(reversed(img.size))])
    processed_outputs = feature_extractor.post_process(outputs, img_size)
    return processed_outputs[0]

def detect_objects(image_input):
    #Extract model and feature extractor
    feature_extractor = AutoFeatureExtractor.from_pretrained("facebook/detr-resnet-50")
    model = DetrForObjectDetection.from_pretrained("facebook/detr-resnet-50")
    
    # if image comes from upload
    if image_input:
        image = image_input
    
    #Make prediction
    processed_outputs = make_prediction(image, feature_extractor, model)
    
    #Visualize prediction
    viz_img = visualize_prediction(image, processed_outputs, 0.7, model.config.id2label)
    
    return viz_img   

# visualization
COLORS = [
    [0.000, 0.447, 0.741],
    [0.850, 0.325, 0.098],
    [0.929, 0.694, 0.125],
    [0.494, 0.184, 0.556],
    [0.466, 0.674, 0.188],
    [0.301, 0.745, 0.933]
]

# Draw the bounding boxes on image.
def fig2img(fig):
    buf = io.BytesIO()
    fig.savefig(buf)
    buf.seek(0)
    img = Image.open(buf)
    return img

# Draw the bounding boxes.
def visualize_prediction(pil_img, output_dict, threshold=0.7, id2label=None):
    keep = output_dict["scores"] > threshold
    boxes = output_dict["boxes"][keep].tolist()
    scores = output_dict["scores"][keep].tolist()
    labels = output_dict["labels"][keep].tolist()
    if id2label is not None:
        labels = [id2label[x] for x in labels]

    plt.figure(figsize=(16, 10))
    plt.imshow(pil_img)
    ax = plt.gca()
    colors = COLORS * 100
    for score, (xmin, ymin, xmax, ymax), label, color in zip(scores, boxes, labels, colors):
        ax.add_patch(plt.Rectangle((xmin, ymin), xmax - xmin, ymax - ymin, fill=False, color=color, linewidth=3))
        ax.text(xmin, ymin, f"{label}: {score:0.2f}", fontsize=15, bbox=dict(facecolor="yellow", alpha=0.5))
    plt.axis("off")
    return fig2img(plt.gcf())


with gr.Blocks(theme=theme, css="footer {visibility: hidden}") as demo:
    with gr.Row():
        with gr.Column():
            image = gr.Image(label="图片", type="pil")
            with gr.Row():
                button = gr.Button("提交", variant="primary")
        box2 = gr.Image(label="图片", shape = (650, 650))

    button.click(fn=detect_objects, inputs=[image], outputs=box2)
    examples = gr.Examples(examples=[['1.jpg'], ['2.jpg'], ['3.jpg']],  inputs=[image], label="例子")


if __name__ == '__main__':
    demo.queue().launch(server_name="0.0.0.0")