setup

README.md

@@ -1,19 +1,32 @@
 ---
-title: Shoplifting Detection
-emoji: 🚀
-colorFrom: red
+title: ShopGuard AI
+emoji: 🛡️
+colorFrom: blue
 colorTo: red
-sdk: docker
-app_port: 8501
-tags:
-- streamlit
+sdk: streamlit
+sdk_version: 1.32.0
+app_file: app.py
 pinned: false
-short_description: Streamlit template space
 ---
 
-# Welcome to Streamlit!
+# 🛡️ ShopGuard AI — Shoplifting Detection System
 
-Edit `/src/streamlit_app.py` to customize this app to your heart's desire. :heart:
+FYP Demo | YOLO11n → MobileNetV2 → Attention LSTM pipeline
 
-If you have any questions, checkout our [documentation](https://docs.streamlit.io) and [community
-forums](https://discuss.streamlit.io).
+## Models
+| Model | HuggingFace Repo | Scene |
+|-------|-----------------|-------|
+| Model A | higsboson/shoplifting_exp_a | General |
+| Model B | higsboson/shoplifting_exp_b | Kitchen |
+| Model C | higsboson/shoplifting_exp_c | Lab |
+
+## How to Use
+1. Select a model from the dropdown
+2. Adjust the decision threshold slider
+3. Upload a short video clip (mp4/avi/mov/mkv)
+4. Click **Run Inference**
+
+## Pipeline
+- **YOLO11n** detects and crops the person in each frame
+- **MobileNetV2** extracts 1280-dim features per frame
+- **Attention LSTM** classifies the sequence as Shoplifting or Normal

requirements.txt

@@ -1,3 +1,6 @@
-altair
-pandas
-streamlit
+streamlit>=1.32.0
+tensorflow>=2.15.0
+numpy>=1.24.0
+opencv-python-headless>=4.8.0
+ultralytics>=8.0.0
+huggingface_hub>=0.21.0

src/README.md

@@ -1,32 +0,0 @@
----
-title: ShopGuard AI
-emoji: 🛡️
-colorFrom: blue
-colorTo: red
-sdk: streamlit
-sdk_version: 1.32.0
-app_file: app.py
-pinned: false
----
-
-# 🛡️ ShopGuard AI — Shoplifting Detection System
-
-FYP Demo | YOLO11n → MobileNetV2 → Attention LSTM pipeline
-
-## Models
-| Model | HuggingFace Repo | Scene |
-|-------|-----------------|-------|
-| Model A | higsboson/shoplifting_exp_a | General |
-| Model B | higsboson/shoplifting_exp_b | Kitchen |
-| Model C | higsboson/shoplifting_exp_c | Lab |
-
-## How to Use
-1. Select a model from the dropdown
-2. Adjust the decision threshold slider
-3. Upload a short video clip (mp4/avi/mov/mkv)
-4. Click **Run Inference**
-
-## Pipeline
-- **YOLO11n** detects and crops the person in each frame
-- **MobileNetV2** extracts 1280-dim features per frame
-- **Attention LSTM** classifies the sequence as Shoplifting or Normal

src/requirements.txt

@@ -1,6 +0,0 @@
-streamlit>=1.32.0
-tensorflow>=2.15.0
-numpy>=1.24.0
-opencv-python-headless>=4.8.0
-ultralytics>=8.0.0
-huggingface_hub>=0.21.0