Update app.py

app.py

@@ -1,3 +1,30 @@
+"""
+# ==============================================================================
+# Vision-Language and Face Recognition Utilities
+# ==============================================================================
+This module provides helper functions, lazy-loading mechanisms, and
+API endpoint wrappers for multimodal inference, face recognition, and
+video scene extraction.
+
+It includes functionality for:
+- Lazy initialization of heavyweight models (vision-language and face models)
+- Image and video preprocessing
+- Multimodal inference with configurable parameters (token limits, temperature)
+- Facial embedding generation
+- Scene extraction from video files
+- Gradio UI components and endpoint definitions for user interaction
+
+All functions and utilities are designed to be:
+- Reusable and cache heavy models to reduce repeated loading
+- Compatible with GPU/CPU execution
+- Stateless and safe to call concurrently from multiple requests
+- Modular, separating model logic from endpoint and UI handling
+
+This module serves as the core interface layer between client-facing
+APIs/UI and the underlying machine learning models.
+# ==============================================================================
+"""
+
 # Standard library
 import json
 import os