app.py

import torch

from transformers import AutoImageProcessor, AutoModelForObjectDetection
#from transformers import pipeline

from PIL import Image
import matplotlib.pyplot as plt
import matplotlib.patches as patches

import io
from random import choice

image_processor_tiny = AutoImageProcessor.from_pretrained("hustvl/yolos-tiny")
model_tiny = AutoModelForObjectDetection.from_pretrained("hustvl/yolos-tiny")

import gradio as gr

COLORS = ["#ff7f7f", "#ff7fbf", "#ff7fff", "#bf7fff",
            "#7f7fff", "#7fbfff", "#7fffff", "#7fffbf",
            "#7fff7f", "#bfff7f", "#ffff7f", "#ffbf7f"]

fdic = {
    "family" : "DejaVu Serif",
    "style" : "normal",
    "size" : 18,
    "color" : "yellow",
    "weight" : "bold"
}


def get_figure(in_pil_img, in_results):
    plt.figure(figsize=(16, 10))
    plt.imshow(in_pil_img)
    ax = plt.gca()

    for score, label, box in zip(in_results["scores"], in_results["labels"], in_results["boxes"]):
        selected_color = choice(COLORS)

        box_int = [i.item() for i in torch.round(box).to(torch.int32)]
        x, y, w, h = box_int[0], box_int[1], box_int[2]-box_int[0], box_int[3]-box_int[1]
        #x, y, w, h = torch.round(box[0]).item(), torch.round(box[1]).item(), torch.round(box[2]-box[0]).item(), torch.round(box[3]-box[1]).item()

        ax.add_patch(plt.Rectangle((x, y), w, h, fill=False, color=selected_color, linewidth=3, alpha=0.8))
        ax.text(x, y, f"{model_tiny.config.id2label[label.item()]}: {round(score.item()*100, 2)}%", fontdict=fdic, alpha=0.8)

    plt.axis("off")

    return plt.gcf()


def infer(in_pil_img, in_threshold=0.9):
    target_sizes = torch.tensor([in_pil_img.size[::-1]])

    inputs = image_processor_tiny(images=in_pil_img, return_tensors="pt")
    outputs = model_tiny(**inputs)

    # convert outputs (bounding boxes and class logits) to COCO API
    results = image_processor_tiny.post_process_object_detection(outputs, threshold=in_threshold, target_sizes=target_sizes)[0]

    figure = get_figure(in_pil_img, results)

    buf = io.BytesIO()
    figure.savefig(buf, bbox_inches='tight')
    buf.seek(0)
    output_pil_img = Image.open(buf)

    return output_pil_img


with gr.Blocks(title="Object Detection") as demo:

    with gr.Row():
        input_image = gr.Image(label="Input image", type="pil")
        output_image = gr.Image(label="Output image with predicted instances", type="pil")

    gr.Examples(['samples/1.jpeg', 'samples/2.JPG'], inputs=input_image)

    threshold = gr.Slider(0, 1.0, value=0.9, label='threshold')

    send_btn = gr.Button("Infer")
    send_btn.click(fn=infer, inputs=[input_image, threshold], outputs=[output_image])

#demo.queue()
demo.launch(debug=True)