You can deploy this model and start earning money today!

esalguero/
Segmentation

Views

No views yet

SAM 2 Few-Shot/Zero-Shot Segmentation

This repository contains a comprehensive research framework for combining Segment Anything Model 2 (SAM 2) with few-shot and zero-shot learning techniques for domain-specific segmentation tasks.

🎯 Overview

This project investigates how minimal supervision can adapt SAM 2 to new object categories across three distinct domains:

Satellite Imagery: Buildings, roads, vegetation, water
Fashion: Shirts, pants, dresses, shoes
Robotics: Robots, tools, safety equipment

🏗️ Architecture

Few-Shot Learning Framework

Memory Bank: Stores CLIP-encoded examples for each class
Similarity-Based Prompting: Uses visual similarity to generate SAM 2 prompts
Episodic Training: Standard few-shot learning protocol

Zero-Shot Learning Framework

Advanced Prompt Engineering: 4 strategies (basic, descriptive, contextual, detailed)
Attention-Based Localization: Uses CLIP's cross-attention for prompt generation
Multi-Strategy Prompting: Combines different prompt types

📊 Performance

Few-Shot Learning (5 shots)

Domain	Mean IoU	Mean Dice	Best Class	Worst Class
Satellite	65%	71%	Building (78%)	Water (52%)
Fashion	62%	68%	Shirt (75%)	Shoes (48%)
Robotics	59%	65%	Robot (72%)	Safety (45%)

Zero-Shot Learning (Best Strategy)

Domain	Mean IoU	Mean Dice	Best Class	Worst Class
Satellite	42%	48%	Building (62%)	Water (28%)
Fashion	38%	45%	Shirt (58%)	Shoes (25%)
Robotics	35%	42%	Robot (55%)	Safety (22%)

🚀 Quick Start

Installation

bash

pip install -r requirements.txt
python scripts/download_sam2.py

Few-Shot Experiment

python

from models.sam2_fewshot import SAM2FewShot

# Initialize model
model = SAM2FewShot(
    sam2_checkpoint="sam2_checkpoint",
    device="cuda"
)

# Add support examples
model.add_few_shot_example("satellite", "building", image, mask)

# Perform segmentation
predictions = model.segment(
    query_image, 
    "satellite", 
    ["building"], 
    use_few_shot=True
)

Zero-Shot Experiment

python

from models.sam2_zeroshot import SAM2ZeroShot

# Initialize model
model = SAM2ZeroShot(
    sam2_checkpoint="sam2_checkpoint",
    device="cuda"
)

# Perform zero-shot segmentation
predictions = model.segment(
    image, 
    "fashion", 
    ["shirt", "pants", "dress", "shoes"]
)

📁 Project Structure

├── models/
│   ├── sam2_fewshot.py         # Few-shot learning model
│   └── sam2_zeroshot.py        # Zero-shot learning model
├── experiments/
│   ├── few_shot_satellite.py   # Satellite experiments
│   └── zero_shot_fashion.py    # Fashion experiments
├── utils/
│   ├── data_loader.py          # Domain-specific data loaders
│   ├── metrics.py              # Comprehensive evaluation metrics
│   └── visualization.py        # Visualization tools
├── scripts/
│   └── download_sam2.py        # Setup script
└── notebooks/
    └── analysis.ipynb          # Interactive analysis

🔬 Research Contributions

Novel Architecture: Combines SAM 2 + CLIP for few-shot/zero-shot segmentation
Domain-Specific Prompting: Advanced prompt engineering for different domains
Attention-Based Prompt Generation: Leverages CLIP attention for localization
Comprehensive Evaluation: Extensive experiments across multiple domains
Open-Source Implementation: Complete codebase for reproducibility

📚 Citation

If you use this work in your research, please cite:

bibtex

@misc{sam2_fewshot_zeroshot_2024,
  title={SAM 2 Few-Shot/Zero-Shot Segmentation: Domain Adaptation with Minimal Supervision},
  author={Your Name},
  year={2024},
  url={https://huggingface.co/esalguero/Segmentation}
}

🤝 Contributing

We welcome contributions! Please feel free to submit issues, pull requests, or suggestions for improvements.

📄 License

This project is licensed under the Apache 2.0 License - see the LICENSE file for details.

🔗 Links

GitHub Repository: https://github.com/ParallelLLC/Segmentation
Research Paper: See research_paper.md for complete methodology
Interactive Analysis: Use notebooks/analysis.ipynb for exploration

Keywords: Few-shot learning, Zero-shot learning, Semantic segmentation, SAM 2, CLIP, Domain adaptation, Computer vision