Scribd
Upload
101 views2 pages

Examples Part 6

This document provides a Python script for performing object detection in images using the Detr model from the Hugging Face Transformers library. It includes code for loading the model, processing images, detecting objects, and visualizing results with bounding boxes. Additionally, it sets up a Gradio interface for users to upload images and view detected objects interactively.

Uploaded by

axen
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as TXT, PDF, TXT or read online on Scribd
101 views2 pages

Examples Part 6

This document provides a Python script for performing object detection in images using the Detr model from the Hugging Face Transformers library. It includes code for loading the model, processing images, detecting objects, and visualizing results with bounding boxes. Additionally, it sets up a Gradio interface for users to upload images and view detected objects interactively.

Uploaded by

axen
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as TXT, PDF, TXT or read online on Scribd

# Example 2.

6
# Object Detection in an Image
!pip install gradio transformers torch matplotlib
import gradio as gr
from transformers import DetrFeatureExtractor, DetrForObjectDetection
import torch
from PIL import Image
import [Link] as plt
from [Link] import Rectangle
import numpy as np

# Load the model and feature extractor


model_name = "facebook/detr-resnet-50"
feature_extractor = DetrFeatureExtractor.from_pretrained(model_name)
model = DetrForObjectDetection.from_pretrained(model_name)

# Function to perform object detection and return an image with boxes


def detect_objects(image):
# Prepare the image for the model
inputs = feature_extractor(images=image, return_tensors="pt")

# Perform inference
outputs = model(**inputs)

# Get the detected boxes and labels


target_sizes = [Link]([[Link][::-1]]) # Model expects (height,
width)
results = feature_extractor.post_process(outputs, target_sizes=target_sizes)[0]

# Visualize the results


[Link](image)
ax = [Link]()

for box, score, label in zip(results["boxes"], results["scores"],


results["labels"]):
if score > 0.9: # Filter by confidence threshold
box = [Link]().numpy()
x, y, w, h = box
rect = Rectangle((x, y), w - x, h - y, linewidth=2, edgecolor='red',
facecolor='none')
ax.add_patch(rect)
label_text = f"{[Link].id2label[[Link]()]}:
{round([Link](), 3)}"
[Link](x, y, label_text, color='white', fontsize=12,
bbox=dict(facecolor='red', alpha=0.5))

[Link]('off')

# Save the image with bounding boxes


[Link]("detected_image.png", bbox_inches='tight')
[Link]()

return [Link]("detected_image.png")

# Gradio interface
interface = [Link](
fn=detect_objects,
inputs=[Link](type="pil"),
outputs=[Link](type="pil"),
title="Object Detection",
description="Upload an image and the model will detect objects with bounding
boxes.",
)

# Launch the app


[Link]()

You might also like