Add application file

app.py

@@ -0,0 +1,53 @@
+import torch
+import requests
+from PIL import Image, ImageDraw, ImageFont
+import gradio as gr
+from transformers import DetrImageProcessor, DetrForObjectDetection
+
+# Load model and processor
+model_name = "facebook/detr-resnet-50"
+processor = DetrImageProcessor.from_pretrained(model_name)
+model = DetrForObjectDetection.from_pretrained(model_name)
+
+# Helper to draw boxes
+def draw_boxes(image, outputs, threshold=0.9):
+    draw = ImageDraw.Draw(image)
+    font = ImageFont.load_default()
+
+    labels = outputs["labels"]
+    boxes = outputs["boxes"]
+    scores = outputs["scores"]
+
+    for score, label, box in zip(scores, labels, boxes):
+        if score >= threshold:
+            box = [round(i, 2) for i in box.tolist()]
+            draw.rectangle(box, outline="red", width=3)
+            text = f"{model.config.id2label[label.item()]}: {round(score.item(), 2)}"
+            draw.text((box[0], box[1]), text, fill="white", font=font)
+
+    return image
+
+# Inference function
+def detect_objects(image):
+    inputs = processor(images=image, return_tensors="pt")
+    with torch.no_grad():
+        outputs = model(**inputs)
+
+    # Post-process results
+    target_sizes = torch.tensor([image.size[::-1]])  # (height, width)
+    results = processor.post_process_object_detection(outputs, target_sizes=target_sizes, threshold=0.0)[0]
+    processed_img = image.copy()
+    result_img = draw_boxes(processed_img, results)
+    return result_img
+
+# Gradio interface
+app = gr.Interface(
+    fn=detect_objects,
+    inputs=gr.Image(type="pil"),
+    outputs=gr.Image(type="pil"),
+    title="Image Detector Agent",
+    description="Upload an image to detect objects using a pretrained DETR model."
+)
+
+if __name__ == "__main__":
+    app.launch()

requirements.txt

@@ -0,0 +1,11 @@
+streamlit
+torch
+torchvision
+matplotlib
+transformers
+pandas
+numpy
+seaborn
+scikit-learn
+opencv-python
+altair<5