Hi HN,
This is Gnarly, an experimental project that blends cellular automata with Deep Dream effects, continuous zooming, and object morphing. The project integrates several ML tools—including Hugging Face’s model downloads, YOLO/EfficientDet for object detection, and Stable Diffusion for creative image generation.
Key challenges I tackled: • Dynamically integrating real-time video/image processing with cellular automata. • Combining multiple deep learning models (InceptionV3 for Deep Dream, YOLO/EfficientDet for object detection, and Stable Diffusion) in a cohesive pipeline. • Optimizing GPU usage and interactive performance with Pygame.
I’d love to hear your thoughts on the technical approach and any suggestions for improvement or future features. Thanks!