Update Gradio interface and add arial.ttf tracked via LFS

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+assets/fonts/arial.ttf filter=lfs diff=lfs merge=lfs -text

.gitignore

@@ -0,0 +1,43 @@
+# Byte-compiled / optimized / DLL files
+__pycache__/
+*.py[cod]
+*.so
+
+# Virtual environment
+venv/
+.env/
+.env.*
+
+# PyCharm
+.idea/
+*.iml
+
+# Model cache (Hugging Face, PyTorch, etc.)
+~/.cache/
+.cache/
+*.ckpt
+*.pt
+*.bin
+*.safetensors
+
+# System files
+.DS_Store
+Thumbs.db
+
+# Logs
+*.log
+
+# Jupyter Notebooks (if any outputs get messy)
+.ipynb_checkpoints/
+
+# Test files or temp
+*.tmp
+*.bak
+
+# Fonts or large assets
+assets/fonts/*.ttf
+
+# Optional: don't track test images
+test_images/
+*.jpg
+*.png

app.py

@@ -1,15 +1,25 @@
 import gradio as gr
+import torch
 from PIL import Image, ImageDraw, ImageFont
 from transformers import DetrImageProcessor, DetrForObjectDetection
-import torch
+from pathlib import Path
+
 
 # Load DETR model and processor from Hugging Face
 model_name = "facebook/detr-resnet-50"
 processor = DetrImageProcessor.from_pretrained(model_name)
 model = DetrForObjectDetection.from_pretrained(model_name)
 
-# Load default font
-font = ImageFont.load_default()
+# Load font
+font_path = Path("assets/fonts/arial.ttf")
+if not font_path.exists():
+    # If the font file does not exist, use the default PIL font
+    print(f"Font file {font_path} not found. Using default font.")
+    font = ImageFont.load_default()
+else:
+    font = ImageFont.truetype(str(font_path), size=100)
+
+print(f"CUDA is available: {torch.cuda.is_available()}")
 
 # Main function: takes an image and returns it with boxes and labels
 def detect_objects(image):
@@ -41,20 +51,24 @@ def detect_objects(image):
         # Set background rectangle for text
         text_background = [
             box[0], box[1] - text_height,
-            box[0] + text_width, box[1]
+                    box[0] + text_width, box[1]
         ]
         draw.rectangle(text_background, fill="black")  # Background
         draw.text((box[0], box[1] - text_height), label_text, fill="white", font=font)
 
     return image_with_boxes
 
-# Gradio interface
-app = gr.Interface(
-    fn=detect_objects,
-    inputs=gr.Image(type="pil"),
-    outputs=gr.Image()
-)
 
-# Run app
+with gr.Blocks() as app:
+    with gr.Row():
+        gr.Markdown("## Object Detection App\nUpload an image to detect objects using Facebook's DETR model.")
+    with gr.Row():
+        input_image = gr.Image(type="pil", label="Input Image")
+        output_image = gr.Image(label="Detected Objects")
+    with gr.Row():
+        button = gr.Button("Detect Objects")
+
+    button.click(fn=detect_objects, inputs=input_image, outputs=output_image)
+
 if __name__ == "__main__":
     app.launch()