<ul><li>A project explores a simplified approach to image generation inspired by the denoising process in diffusion models.</li><li>The project uses the CIFAR-10 dataset with 100 classes of labeled images for training a U-Net architecture for image generation tasks.</li><li>The U-Net model is trained to reconstruct clean images from noisy inputs using mean squared error loss and the Adam optimizer.</li><li>The project methodology includes stages such as corruption, reconstruction, and image generation, showcasing promising results despite computational constraints.</li></ul>

Image Generation via Diffusion

Discover more