# CSE5519 Advances in Computer Vision (Topic E: 2023: Deep Learning for Geometric Computer Vision)

## Guided Depth Super-Resolution by Deep Anisotropic Diffusion

[link to paper](https://openaccess.thecvf.com/content/CVPR2023/papers/Metzger_Guided_Depth_Super-Resolution_by_Deep_Anisotropic_Diffusion_CVPR_2023_paper.pdf)

Combining anisotropic diffusion (discretized version of the heat equation) with deep feature learning (CNN).

For upscaling.

>[!TIP]
>
> How the anisotropic diffusion adjustment steps are used to guide the super-resolution process? How the deep feature learning is used during the process?