Update app.py

app.py

@@ -1,62 +1,41 @@
-# Install and update the necessary libraries
-#import logging
-#import sys
-from PIL import Image
-from transformers import pipeline
-import os
 import gradio as gr
+from transformers import pipeline
+from helper import render_results_in_image, summarize_predictions_natural_language, ignore_warnings
 
-# Suppress non-critical log messages
-from transformers.utils import logging
-logging.set_verbosity_error()
-
-#from helper import ignore_warnings
-#ignore_warnings()
-#import sys
-#sys.path.append("./Object-Detection/helper.py")  # Adjust the path as necessary to point to the directory where helper.py is located
-
-#from Object-Detection.helper import ignore_warnings
-#from helper import ignore_warnings
-from helper import *
+# Suppress non-critical warnings
 ignore_warnings()
 
-# Import the pipeline function from the transformers library
-#from transformers import pipeline
+# Load DETR object detection pipeline with lower threshold
+pipe = pipeline("object-detection", model="facebook/detr-resnet-50", threshold=0.3)
 
-# Set up the object detection pipeline.
-pipe = pipeline("object-detection", model="facebook/detr-resnet-50") 
-
-#sys.path.append("./Open_Source_Models_with_Hugging_Face/Object_Detection/helper.py")  # Adjust the path as necessary to point to the directory where helper.py is located
-
-#from Object-Detection.helper import import load_image_from_url, render_results_in_image
-#from helper import load_image_from_url, render_results_in_image
+# Function to get predictions and process image + summary
+def get_pipeline_prediction(pil_image):
+    pipeline_output = pipe(pil_image)
 
+    if not pipeline_output:
+        return pil_image, "No objects detected."
 
-# Load the image from a file
-#raw_image = Image.open('./Open_Source_Models_with_Hugging_Face/Object_Detection/kittens.jpeg')
-# Resize the image 
-#raw_image.resize((569, 491))
+    processed_image = render_results_in_image(pil_image, pipeline_output)
+    summary = summarize_predictions_natural_language(pipeline_output)
 
-#pipeline_output = pipe(raw_image)
-#processed_image = render_results_in_image(raw_image, pipeline_output)
-#processed_image
+    return processed_image, summary
 
-# gradio interface 
-def get_pipeline_prediction(pil_image):
-    pipeline_output = pipe(pil_image)
-    processed_image = render_results_in_image(pil_image, pipeline_output)
-    return processed_image
 
+# Gradio interface
 iface = gr.Interface(
-  fn=get_pipeline_prediction,
-  inputs=gr.Image(label="Input image", 
-                  type="pil"),
-  outputs=gr.Image(label="Output image with predicted instances",
-                   type="pil")
+    fn=get_pipeline_prediction,
+    inputs=gr.Image(label="Upload an Image", type="pil"),
+    outputs=[
+        gr.Image(label="Image with Detected Objects", type="pil"),
+        gr.Textbox(label="Summary of Detected Objects")
+    ],
+    title="Object Detection with DETR",
+    description="Upload an image to detect objects. Bounding boxes will be drawn and a natural language summary will be provided.",
+    flagging_mode="never"
 )
 
-iface.launch()
-
+if __name__ == "__main__":
+    iface.launch()