1. Handcrafted Image Representations
- Implemented three different image representations for MNIST dataset:
- Raw Pixels
- Histogram of Gradients
- Local Binary Patterns
2. Image Denoising with a Deep Image Prior
- Denoising images with
- AutoEncoder with Batch Normalization and Bilinear Upsampling
- Deeper Architectures, those with Skip Connections, as well as Averaging the Results across iterations.