FossilFrame

• 

  Landing Page

• 

  FossilFrame in action

Inspiration

We were inspired by the idea that a lot of Earth’s history is literally hidden due to physical limitations in imaging. When we learned about the diffraction limit in microscopes, it felt like there was so much untapped data just sitting there. That led us to explore whether combining quantum concepts with AI could help us explore beyond those limits.

What it does

FossilFrame is a quantum-inspired imaging system that enhances fossil images by recovering details lost in traditional microscopy. It takes blurred microscope-like images and reconstructs clearer versions using quantum simulation and AI, helping reveal structures that are otherwise invisible.

How we built it

We built the project as a modular pipeline. First, we simulated classical optical blur using OpenCV. Then we used Qiskit to generate quantum-inspired probability distributions that mimic hidden structural information. Finally, we applied AI models using TensorFlow/PyTorch to reconstruct enhanced images. Everything was integrated into a simple interface using Streamlit.

Challenges we ran into

One major challenge was connecting concepts from completely different domains — quantum computing, image processing, and machine learning — into a single working pipeline. We also struggled with making the outputs meaningful and not just visually different. Tuning the reconstruction and ensuring consistency in results took a lot of iteration.

Accomplishments that we're proud of

We’re proud that we were able to successfully build a working hybrid system that combines quantum simulation with AI in a meaningful way. Getting a clear visual difference between classical and reconstructed outputs was a big win for us. Also, making such a complex idea understandable and demo-friendly felt like a major achievement.

What we learned

We learned how to think across domains instead of sticking to just one field. This project helped us understand quantum computing concepts more practically, along with improving our skills in computer vision and ML. More importantly, we learned how to break down a complex idea and actually implement it step by step.

What's next for FossilFrame

In the future, we want to improve the accuracy of the reconstruction and explore using real quantum hardware instead of simulations. We also see potential in applying this approach to medical imaging and other scientific fields where capturing fine details is critical.

Built With

• opencv
• python
• pytorch
• qiskit
• tensorflow