app.py

import os

import torch
import torch.nn.functional as F
import torchvision.transforms as T
from mmdet.apis import init_detector, inference_detector, show_result_pyplot
import mmcv

import gradio as gr
from huggingface_hub import hf_hub_download

# Device on which to run the model
# Set to cuda to load on GPU
device = "cpu"
checkpoint_file = hf_hub_download(repo_id="Andy1621/uniformer", filename="mask_rcnn_3x_ms_hybrid_small.pth")
config_file = './exp/mask_rcnn_3x_ms_hybrid_small/config.py'
# init detector
# build the model from a config file and a checkpoint file
model = init_detector(config_file, checkpoint_file, device='cpu')


def set_example_image(example: list) -> dict:
    return gr.Image.update(value=example[0])


def inference(img):
    result = inference_detector(model, img)
    res_img = show_result_pyplot(model, img, result).numpy()
    return res_img


demo = gr.Blocks()
with demo:
    gr.Markdown(
        """
        # UniFormer-S
        Gradio demo for <a href='https://github.com/Sense-X/UniFormer' target='_blank'>UniFormer</a>: To use it, simply upload your image, or click one of the examples to load them. Read more at the links below.
        """
    )

    with gr.Box():
        with gr.Row():
                with gr.Column():
                    with gr.Row():
                        input_image = gr.Image(label='Input Image', type='numpy')
                    with gr.Row():
                        submit_button = gr.Button('Submit')
                with gr.Column():
                    res_image = gr.Image(type='numpy', label='Detection Resutls')
        with gr.Row():
            example_images = gr.Dataset(components=[input_image], samples=[['demo.jpg']])

    gr.Markdown(
        """
        <p style='text-align: center'><a href='https://arxiv.org/abs/2201.09450' target='_blank'>UniFormer: Unifying Convolution and Self-attention for Visual Recognition</a> | <a href='https://github.com/Sense-X/UniFormer' target='_blank'>Github Repo</a></p>
        """
    )

    submit_button.click(fn=inference, inputs=input_image, outputs=res_image)
    example_images.click(fn=set_example_image, inputs=example_images, outputs=example_images.components)

demo.launch(enable_queue=True)