Spaces:
Running
Running
import os | |
import subprocess | |
# Clone the yolov5 repository and install its requirements | |
if not os.path.exists('yolov5'): | |
subprocess.run(['git', 'clone', 'https://github.com/ultralytics/yolov5'], check=True) | |
subprocess.run(['pip', 'install', '-r', 'yolov5/requirements.txt'], check=True) | |
import torch | |
import torchvision | |
from torchvision.transforms import functional as F | |
from PIL import Image | |
import cv2 | |
import gradio as gr | |
import numpy as np | |
from yolov5.models.yolo import Model | |
from yolov5.utils.general import non_max_suppression | |
device = torch.device('cuda' if torch.cuda.is_available() else 'cpu') | |
print(f"Using device: {device}") | |
model = torch.hub.load('ultralytics/yolov5', 'yolov5s', pretrained=True).to(device) | |
model.eval() | |
print("Model loaded successfully") | |
def preprocess_image(image): | |
try: | |
image = Image.fromarray(image) # Convert numpy array to PIL Image | |
image_tensor = F.to_tensor(image).unsqueeze(0).to(device) | |
print(f"Preprocessed image tensor: {image_tensor.shape}") | |
return image_tensor | |
except Exception as e: | |
print(f"Error in preprocessing image: {e}") | |
return None | |
def draw_boxes(image, outputs, threshold=0.3): | |
try: | |
image = cv2.cvtColor(image, cv2.COLOR_RGB2BGR) | |
h, w, _ = image.shape | |
for box in outputs: | |
if box is not None: | |
x1, y1, x2, y2, score, label = box[:6] | |
if score > threshold: | |
x1, y1, x2, y2 = int(x1), int(y1), int(x2), int(y2) | |
cv2.rectangle(image, (x1, y1), (x2, y2), (255, 0, 0), 2) | |
text = f"{model.names[int(label)]:s}: {score:.2f}" | |
cv2.putText(image, text, (x1, y1 - 5), cv2.FONT_HERSHEY_SIMPLEX, 0.5, (255, 0, 0), 2) | |
return cv2.cvtColor(image, cv2.COLOR_BGR2RGB) | |
except Exception as e: | |
print(f"Error in drawing boxes: {e}") | |
return image | |
def detect_objects(image): | |
image_tensor = preprocess_image(image) | |
if image_tensor is None: | |
return image | |
try: | |
outputs = model(image_tensor)[0] # Get the first element of the output | |
print(f"Model raw outputs: {outputs}") | |
outputs = non_max_suppression(outputs, conf_thres=0.25, iou_thres=0.45)[0] # Apply NMS | |
if outputs is None or len(outputs) == 0: | |
print("No objects detected.") | |
return image | |
print(f"Filtered outputs: {outputs}") | |
result_image = draw_boxes(image, outputs.cpu().numpy()) | |
return result_image | |
except Exception as e: | |
print(f"Error in detecting objects: {e}") | |
return image | |
iface = gr.Interface( | |
fn=detect_objects, | |
inputs=gr.Image(type="numpy"), | |
outputs=gr.Image(type="numpy"), | |
title="YOLOv5 Object Detection", | |
description="Upload an image to detect objects using the YOLOv5 model." | |
) | |
if __name__ == "__main__": | |
iface.launch() | |