18637
Overview
An AI algorithm using a pre-trained Vision Transformer (ViT) to transfer one image’s appearance onto another’s structure, creating realistic chimeric images automatically and efficiently without extra input.
Applications
- A feature for AI systems that can create realistic images with different appearances
- Add-on feature for photo editing software to modify HD photos
- Real-time photo modification tool for social media applications
Differentiation
- Fast and precise image editing
- Creation of high-definition, realistic chimeric images
- Broad applicability to various objects and scenes
Development Status
The ViT algorithm is developed, demonstrated, and published.
The algorithm combines a source image's structure with a target image's appearance to generate a new, semantically consistent image.
Reference
Patent Status:
USA Granted: 12,282,696
Contact for more information
