Deep Learning for Computer Vision

• Final Project: Multi-concept Text-to-Image Personalization
• 3D novel view synthesis (3D gaussian splatting)
• Zero-shot image captioning with LLaVA
• Image captioning using PEFT on vision language models
• Conditional diffusion models (DDPM)
• Generating human faces (DDIM)
• Text-to-image personalization (DDPM, stable diffusion)
• Semantic segmentation (FCN, DeepLabV3, SAM)
• Self-supervised learning (ResNet)

Principles and Applications of Digital Image Processing

• Image fusion with discrete wavelet transform (DWT)
• Removing motion blur with Wiener filter
• Fourier transform and inverse fourier transform

Introduction to Artificial Intelligence

• 3D scene reconstruction using neural radiance fields (NeRF)
• Multimodal representation learning with ImageBind