---
title: Final Project Object Localization
emoji: 🏆
colorFrom: indigo
colorTo: gray
sdk: gradio
sdk_version: 5.29.0
app_file: app.py
pinned: false
license: cc
---

Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

---

# Object Localization — Bounding Box Prediction

**Authors:** Elizabeth Wangley and Henry Wang  
**Course:** CSCI 4750 - Machine Learning  
**Date:** May 12, 2025  
**Paper:** [Read the full paper](https://docs.google.com/document/d/1j6jLVY_zNitulqg8Gps824ZlW59BBOY2LYh2olP-B4k/edit?usp=sharing)  
**Live Demo:** [Hugging Face Spaces](https://huggingface.co/spaces/MLFinalProject/Final_Project_Object_Localization)

---

## Overview

A Gradio web app that predicts bounding boxes around a stop sign in images using two ML models:

- **Linear Regression** — Four separate models, one per coordinate (`x`, `y`, `width`, `height`)
- **MLP (Neural Network)** — Single model predicting all four coordinates simultaneously

Images are resized to 64×64, converted to grayscale, and flattened as model input. The app displays the predicted box in **red** and the ground truth in **green** (when available).

---

## Results

| Metric    | Linear Regression | MLP   |
| --------- | ----------------- | ----- |
| Avg. RMSE | ~7                | ~3    |
| Mean IoU  | 0.581             | 0.778 |

The MLP outperformed Linear Regression and was selected for deployment.

---

## Usage

1. Upload an image
2. Select a model — **MLP** or **Linear Regression**
3. View the predicted bounding box vs. ground truth

---

## Run Locally

```bash
pip install -r requirements.txt
python app.py
```

Open `http://127.0.0.1:7860` in your browser.

---

## Project Structure

```
├── app.py
├── image_annotation.csv
├── mlp_bbox_model.pkl
├── linear_model_x.pkl
├── linear_model_y.pkl
├── linear_model_w.pkl
├── linear_model_h.pkl
├── requirements.txt
└── README.md
```

---

## Installation & Running Locally

```bash
git clone https://github.com/your-username/Final_Project_Object_Localization.git
cd Final_Project_Object_Localization
pip install -r requirements.txt
python app.py
```

Then open `http://127.0.0.1:7860` in your browser.

---

## Requirements

- Python 3.8+
- gradio
- numpy
- pandas
- scikit-learn
- scikit-image
- matplotlib
- joblib

---

## Future Work

- Explore additional model architectures and do a more thorough comparison of traditional ML vs. deep learning in terms of resource cost vs. accuracy.
- Test a broader set of hyperparameters with stronger computing resources.
- Investigate alternative hyperparameter tuning strategies (e.g., random search, Bayesian optimization) for efficiency.
- Evaluate model robustness on images where the stop sign is partially visible or occluded.

---

## References

- Géron, Aurélien. _Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow_. O'Reilly Media, 2022.
- "Hyperparameter Tuning in Linear Regression." GeeksforGeeks, July 1, 2024. https://www.geeksforgeeks.org/hyperparameter-tuning-in-linear-regression/
- Nayda, Yuriy. "Object Detection Technology: How It Works and Where It's Used." SmartTek Solutions, Jan. 11, 2023. https://smarttek.solutions/blog/object-detection-technology/