Upload 3 files

Chicken_Rabbit_Detection.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:37801d43221f81f8116d09951d6303d597ed9a1a0169e2b5a2bd43f3c53f1e1d
+size 19195546

app.py

@@ -0,0 +1,98 @@
+from ultralytics import YOLO
+from PIL import Image
+import gradio as gr
+from huggingface_hub import snapshot_download
+from huggingface_hub import hf_hub_download
+import os
+import tempfile
+
+MODEL_ID = "Chicken_Rabbit_Detection.pt"
+REPO_ID = "ITI121-25S2/1552444F"
+
+def load_model(repo_id):
+    # Download the single model file from Hugging Face model repository
+
+    path = hf_hub_download(repo_id=REPO_ID, filename=MODEL_ID)
+
+    print(path)
+    detection_model = YOLO(path, task='detect')
+    return detection_model
+
+
+def predict(input_data):
+    """
+    Process either an image (PIL Image) or a video file path.
+    Returns processed image or video file path.
+    """
+    # Check if input is a PIL Image (image upload)
+    if isinstance(input_data, Image.Image):
+        # Image processing
+        source = input_data
+        result = detection_model.predict(source, conf=0.5, iou=0.6)
+        img_bgr = result[0].plot()
+        out_pilimg = Image.fromarray(img_bgr[..., ::-1])  # RGB-order PIL image
+        return out_pilimg
+    
+    # Otherwise, it's a video file path (string)
+    elif isinstance(input_data, str) and os.path.isfile(input_data):
+        # Video processing
+        # Get the actual temp directory path (resolves /var to /private/var on macOS)
+        temp_dir = os.path.realpath(tempfile.gettempdir())
+        
+        # Process video with YOLO
+        result = detection_model.predict(
+            source=input_data,
+            conf=0.75,
+            iou=0.6,
+            save=True,
+            project=temp_dir,
+            name='gradio_video_output',
+            exist_ok=True
+        )
+        
+        # Find the output video file
+        # YOLO saves to runs/detect/gradio_video_output/ by default
+        save_dir = result[0].save_dir
+        # detect_dir = os.path.join(temp_dir, 'runs', 'detect', 'gradio_video_output')
+        # Resolve any symlinks to get the actual path
+        # detect_dir = os.path.realpath(detect_dir) if os.path.exists(detect_dir) else detect_dir
+        
+        if os.path.exists(save_dir):
+            # Find the video file in the output directory
+            video_files = [f for f in os.listdir(save_dir) if f.endswith(('.mp4', '.avi', '.mov', '.mkv'))]
+            if video_files:
+                output_path = os.path.join(save_dir, video_files[0])
+                # Resolve the final path to handle any symlinks
+                output_path = os.path.realpath(output_path)
+                return output_path
+        
+        # Fallback: return input if processing failed
+        return input_data
+    
+    else:
+        raise ValueError("Input must be either a PIL Image or a video file path")
+
+detection_model = load_model(REPO_ID)
+
+# Create interface with both image and video inputs
+with gr.Blocks() as demo:
+    gr.Markdown("# 🐔🐇 Chicken & Rabbit Detection")
+    gr.Markdown(
+        "Upload an image or video file to detect **chickens and rabbits** "
+        "using a trained YOLO object detection model."
+    )
+
+    with gr.Tabs():
+        with gr.TabItem("Image"):
+            image_input = gr.Image(type="pil", label="Upload Image")
+            image_output = gr.Image(type="pil", label="Detected Chickens & Rabbits")
+            image_btn = gr.Button("Detect")
+            image_btn.click(fn=predict, inputs=image_input, outputs=image_output)
+
+        with gr.TabItem("Video"):
+            video_input = gr.Video(label="Upload Video")
+            video_output = gr.Video(label="Detected Chickens & Rabbits")
+            video_btn = gr.Button("Detect")
+            video_btn.click(fn=predict, inputs=video_input, outputs=video_output)
+
+demo.launch(share=True)

requirements.txt

@@ -0,0 +1,3 @@
+ultralytics
+huggingface_hub
+gradio