A group of students worked together on the 'Bringing Photos Back To Life' project, which restores old photographs that have gone through severe degradation. Using layers of technology, the project developed a solution that learns the visuals through supervised learning techniques and artificial intelligence.
It measures the domain gap between the old photos and synthetic images through a novel triplet domain translation. Designing a global branch with partially nonlocal block targeting as well as the structured defects including dust, scratches, and measuring noise and blurriness. The method trains two VAE, which are variational auto encoders to two latent spaces using the paired data of old and clean photos to tackle the degradations in the old photos.
Image Credit: Bringing Old Photos Back to Life, Bo Zhang
Key Themes Behind This Trend
- Restorative Photography
- Utilizing AI and supervised learning techniques to restore degraded old photographs.
- Image Domain Translation
- Developing triplet domain translation to bridge the gap between old photos and synthetic images.
- Variational Auto Encoders (VAE)
- Training VAEs to address degradations like dust, scratches, noise, and blurriness in old photos.
Where This Applies
- Photography
- Leveraging AI and image restoration techniques to offer photo restoration services.
- Artificial Intelligence
- Applying AI to improve the quality and visual appeal of old photographs.
- Technology
- Developing innovative solutions to restore and revive degraded images using advanced techniques like VAE.
