Update app.py

app.py

@@ -1,89 +1,67 @@
+import gradio as gr
 from transformers import DetrImageProcessor, DetrForObjectDetection
 import torch
-from PIL import Image, ImageDraw, ImageFont  # Import ImageFont
-import gradio as gr
-import requests
+from PIL import Image, ImageDraw
 import random
 
 def detect_objects(image):
-    # Load the pre-trained DETR model
     processor = DetrImageProcessor.from_pretrained("facebook/detr-resnet-50")
     model = DetrForObjectDetection.from_pretrained("facebook/detr-resnet-50")
 
     inputs = processor(images=image, return_tensors="pt")
     outputs = model(**inputs)
 
-    # convert outputs (bounding boxes and class logits) to COCO API
-    # let's only keep detections with score > 0.9
     target_sizes = torch.tensor([image.size[::-1]])
     results = processor.post_process_object_detection(outputs, target_sizes=target_sizes, threshold=0.9)[0]
 
-    # Draw bounding boxes and labels on the image
     draw = ImageDraw.Draw(image)
+    labels = []
     for i, (score, label, box) in enumerate(zip(results["scores"], results["labels"], results["boxes"])):
         box = [round(i, 2) for i in box.tolist()]
         color = (random.randint(0, 255), random.randint(0, 255), random.randint(0, 255))
         draw.rectangle(box, outline=color, width=3)
         label_text = f"{model.config.id2label[label.item()]}: {round(score.item(), 2)}"
-        # Larger and bolder font
         draw.text((box[0], box[1]), label_text, fill=color,)
+        labels.append(label_text)
 
-    return image
-
-def detect_labels(image):
-    # Load the pre-trained DETR model
-    processor = DetrImageProcessor.from_pretrained("facebook/detr-resnet-50")
-    model = DetrForObjectDetection.from_pretrained("facebook/detr-resnet-50")
-
-    inputs = processor(images=image, return_tensors="pt")
-    outputs = model(**inputs)
+    return image, labels
 
-    # convert outputs (bounding boxes and class logits) to COCO API
-    # let's only keep detections with score > 0.9
-    target_sizes = torch.tensor([image.size[::-1]])
-    results = processor.post_process_object_detection(outputs, target_sizes=target_sizes, threshold=0.9)[0]
-
-    labels = []
-    for label_id in results["labels"]:
-        labels.append(model.config.id2label[label_id.item()])
-
-    return labels
-
-def upload_image_with_boxes(file):
+def upload_image(file):
     image = Image.open(file.name)
-    image_with_boxes = detect_objects(image)
-    return image_with_boxes
+    image_with_boxes, labels = detect_objects(image)
+    return image_with_boxes, labels
 
-def upload_image_with_labels(file):
-    image = Image.open(file.name)
-    labels = detect_labels(image)
-    return ", ".join(labels)
+def show_labels(labels):
+    return "\n".join(labels)
 
-iface_switch = gr.Interface(
-    fn=upload_image_with_boxes,
+# Interface to display the image with bounding boxes
+iface_objects = gr.Interface(
+    fn=upload_image,
     inputs="file",
-    outputs="image",
-    title="Object Detection with Boxes",
-    description="Upload an image and detect objects using DETR model. Tap the image to switch to labels view.",
+    outputs=["image", "text"],
+    title="Object Detection",
+    description="Upload an image and detect objects using DETR model.",
     allow_flagging=False
 )
 
+# Interface to display the detected labels
 iface_labels = gr.Interface(
-    fn=upload_image_with_labels,
-    inputs="file",
+    fn=show_labels,
+    inputs="text",
     outputs="text",
-    title="Detected Object Labels",
-    description="Upload an image and get the detected object labels using DETR model. Tap the text to switch back to object detection with boxes view.",
+    title="Detected Labels",
+    description="Displays the labels detected in the uploaded image.",
     allow_flagging=False
 )
 
-iface_tapped = gr.Interface(
-    [iface_switch, iface_labels],
-    inputs="file",
-    outputs="image",
-    title="Object Detection and Labels",
-    description="Upload an image and switch between object detection with boxes and displaying detected object labels as text.",
+# Combine interfaces with a tapped interface
+interface = gr.Interface(
+    [iface_objects, iface_labels],
+    inputs="text",
+    outputs="text",
+    title="Object Detection with Labels",
+    description="Upload an image and view detected objects and labels.",
     allow_flagging=False
 )
 
-iface_tapped.launch()
+interface.launch()