Vision-Aid

Vision-Aid is an AI-powered tool designed to generate audio descriptions of video content, enhancing accessibility for visually impaired individuals. By analyzing video frames, it provides contextual audio narratives, enabling users to comprehend visual media through sound.

---

🎯 Features

• Video Frame Analysis: Utilizes deep learning models to extract meaningful features from video frames.
• Contextual Audio Narration: Translates visual elements into coherent audio descriptions.
• Pretrained Models: Includes encoder and decoder models for efficient processing.
• Customizable Training: Offers scripts to train models on custom datasets.

---

🗂️ Project Structure

Vision-Aid/
├── Train_data/                 # Directory containing training data
├── IDs.txt                     # List of video IDs used for training/testing
├── Test.py                     # Script to test the model on new data
├── decoder_model_weights.h5    # Pretrained decoder model weights
├── encoder_model.h5            # Pretrained encoder model
├── featureExtractor.py         # Script to extract features from video frames
├── feature_dict2.pickle        # Serialized feature dictionary
├── model_train.py              # Script to train the model
├── tokenizer/                  # Directory containing tokenizer data

---

🚀 Getting Started

Prerequisites

• Python 3.6 or higher
• Required Python packages (see below)

Installation

1. Clone the repository:

git clone https://github.com/Qadeer2syed/Vision-Aid.git
cd Vision-Aid

2. Install dependencies:

It's recommended to use a virtual environment.

python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate
pip install -r requirements.txt

Note: Ensure that requirements.txt contains all necessary packages.

---

🧪 Usage

Feature Extraction

Extract features from video frames using the encoder model:

python featureExtractor.py --input_dir path_to_videos --output_file feature_dict2.pickle

Model Training

Train the model with your dataset:

python model_train.py --features feature_dict2.pickle --ids IDs.txt

Testing

Generate audio descriptions for new videos:

python Test.py --video path_to_video

Ensure that the decoder_model_weights.h5 and encoder_model.h5 files are present in the project directory.

---

📁 Dataset

The Train_data/ directory should contain subdirectories or files representing the training videos. The IDs.txt file should list the identifiers corresponding to the training data.

---