kasper-boy's picture
Update app.py
5a46b3c verified
import torch
from transformers import DetrImageProcessor, DetrForObjectDetection
from PIL import Image
import gradio as gr
import matplotlib.pyplot as plt
import matplotlib.patches as patches
import io
# Load the processor and model
processor = DetrImageProcessor.from_pretrained('facebook/detr-resnet-101')
model = DetrForObjectDetection.from_pretrained('facebook/detr-resnet-101')
def object_detection(image):
# Preprocess the image
inputs = processor(images=image, return_tensors="pt")
# Perform object detection
outputs = model(**inputs)
# Extract bounding boxes and labels
target_sizes = torch.tensor([image.size[::-1]])
results = processor.post_process_object_detection(outputs, target_sizes=target_sizes, threshold=0.9)[0]
# Plot the image with bounding boxes
plt.figure(figsize=(16, 10))
plt.imshow(image)
ax = plt.gca()
for score, label, box in zip(results["scores"], results["labels"], results["boxes"]):
box = [round(i, 2) for i in box.tolist()]
xmin, ymin, xmax, ymax = box
width, height = xmax - xmin, ymax - ymin
ax.add_patch(plt.Rectangle((xmin, ymin), width, height, fill=False, color='red', linewidth=3))
text = f'{model.config.id2label[label.item()]}: {round(score.item(), 3)}'
ax.text(xmin, ymin, text, fontsize=15, bbox=dict(facecolor='yellow', alpha=0.5))
plt.axis('off')
# Save the plot to an image buffer
buf = io.BytesIO()
plt.savefig(buf, format='png')
buf.seek(0)
plt.close()
# Convert buffer to an Image object
result_image = Image.open(buf)
return result_image
# Define the Gradio interface
demo = gr.Interface(
fn=object_detection,
inputs=gr.Image(type="pil", label="Upload an Image"),
outputs=gr.Image(type="pil", label="Detected Objects"),
title="Object Detection with DETR (ResNet-101)",
description="Upload an image and get object detection results using the DETR model with a ResNet-101 backbone.",
)
# Launch the Gradio interface
if __name__ == "__main__":
demo.launch()