File size: 5,882 Bytes
95787c5 d6f8f6f 95787c5 f13b6ca 95787c5 dcd8a6a 95787c5 dcd8a6a acdc5b0 95787c5 acdc5b0 95787c5 acdc5b0 95787c5 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 |
import os
import socket
import gradio as gr
import numpy as np
from PIL import Image, ImageDraw
from pathlib import Path
from loguru import logger
import cv2
import torch
import time
import base64
import requests
import json
# API for inferences
DL4EO_API_URL = "https://dl4eo--groundingdino-predict.modal.run"
# Auth Token to access API
DL4EO_API_KEY = os.environ['DL4EO_API_KEY']
# width of the boxes on image
LINE_WIDTH = 2
# Check Gradio modules version
logger.info(f"Gradio version: {gr.__version__}")
# Define the inference function
def predict_image(image, text_prompt, box_threshold, text_threshold):
# Resize the image to the new size
#image = image.resize((image.size[0] * 2, image.size[1] * 2))
if isinstance(image, Image.Image):
img = np.array(image)
if not isinstance(img, np.ndarray) or len(img.shape) != 3 or img.shape[2] != 3:
raise BaseException("predit_image(): input 'img' shoud be single RGB image in PIL or Numpy array format.")
#width, height = img.shape[0], img.shape[1]
# Encode the image data as base64
image_base64 = base64.b64encode(np.ascontiguousarray(img)).decode()
# Create a dictionary representing the JSON payload
payload = {
'image': image_base64,
'shape': img.shape,
'text_prompt': text_prompt,
'box_threshold': box_threshold,
'text_threshold': text_threshold,
}
headers = {
'Authorization': 'Bearer ' + DL4EO_API_KEY,
'Content-Type': 'application/json' # Adjust the content type as needed
}
# Send the POST request to the API endpoint with the image file as binary payload
response = requests.post(DL4EO_API_URL, json=payload, headers=headers)
# Check the response status
if response.status_code != 200:
raise Exception(
f"Received status code={response.status_code} in inference API: {response.text}"
)
json_data = json.loads(response.content)
duration = json_data['duration']
boxes = json_data['boxes']
# drow boxes on image
draw = ImageDraw.Draw(image)
for box in boxes:
left, top, right, bottom = box
if left <= 0: left = -LINE_WIDTH
if top <= 0: top = top - LINE_WIDTH
if right >= img.shape[0] - 1: right = img.shape[0] - 1 + LINE_WIDTH
if bottom >= img.shape[1] - 1: bottom = img.shape[1] - 1 + LINE_WIDTH
draw.rectangle([left, top, right, bottom], outline="red", width=LINE_WIDTH)
return image, str(image.size), len(boxes), duration
# Define example images and their true labels for users to choose from
example_data = [
["./demo/Pleiades_Neo_Tucson_USA.jpg", 'plane', 0.24, 0.24],
["./demo/Pleiades_Neo_Tucson_USA.jpg", 'building', 0.24, 0.24],
#["./demo/Pleiades_Neo_Tucson_USA.jpg", 'tree', 0.24, 0.24],
#["./demo/two-dogs-with-a-stick.jpg", "dog", 0.25, 0.25],
#["./demo/airport01.jpg", "aircraft", 0.25, 0.25],
#["./demo/SPOT_Storage.jpg", "storage", 0.25, 0.25],
#["./demo/Satellite_Image_Marina_New_Zealand.jpg", "ship", 0.25, 0.25],
["./demo/Pleiades_HD15_Miami_Marina.jpg", "motorboat", 0.3, 0.0],
["./demo/Pleiades_HD15_Miami_Marina.jpg", "palm tree", 0.15, 0.3],
["./demo/Pleiades_HD15_Miami_Marina.jpg", "building", 0.3, 0.0],
]
# Define CSS for some elements
css = """
.image-preview {
height: 820px !important;
width: 800px !important;
}
"""
TITLE = "Open detection on optical satellite images"
# Define the Gradio Interface
demo = gr.Blocks(title=TITLE, css=css).queue()
with demo:
gr.Markdown(f"<h1><center>{TITLE}<center><h1>")
with gr.Row():
with gr.Column(scale=0):
input_image = gr.Image(type="pil", interactive=True, scale=1)
text_prompt = gr.Textbox(label="Text prompt")
run_button = gr.Button(value="Run", scale=0)
with gr.Accordion("Advanced options", open=True):
box_threshold = gr.Slider(label="Box threshold", minimum=0.0, maximum=1.0, value=0.24, step=0.01)
text_threshold = gr.Slider(label="Text threshold", minimum=0.0, maximum=1.0, value=0.24, step=0.01)
dimensions = gr.Textbox(label="Image size", interactive=False)
detections = gr.Number(label="Predicted objects", interactive=False)
stopwatch = gr.Number(label="Execution time (sec.)", interactive=False, precision=3)
with gr.Column(scale=2):
output_image = gr.Image(type="pil", elem_classes='image-preview', interactive=False, width=800, height=800)
run_button.click(fn=predict_image, inputs=[input_image, text_prompt, box_threshold, text_threshold], outputs=[output_image, dimensions, detections, stopwatch])
gr.Examples(
examples=example_data,
inputs = [input_image, text_prompt, box_threshold, text_threshold],
outputs = [output_image, dimensions, detections, stopwatch],
fn=predict_image,
cache_examples=True,
label='Try these images!'
)
gr.Markdown("<p>This demo is provided by <a href='https://www.linkedin.com/in/faudi/'>Jeff Faudi</a> \
and <a href='https://www.dl4eo.com/'>DL4EO</a>. The demonstration images are Pléiades \
images provided by CNES with distribution by Airbus DS. The model architecture and weights \
are provided <a href='https://github.com/IDEA-Research/GroundingDINO'>Grounding DINO</a>. \
The model has not been trained specifically on satellite imagery and should be finetuned for this task. \
This is for demonstration only. Please contact <a href='mailto:[email protected]'>me</a> \
for more information on how you could get access to a commercial model or API. </p>")
demo.launch(
inline=False,
show_api=False,
debug=False
)
|