Monocular Depth Estimation: a Review of the 2022 State of the Art

We compare five monocular depth estimation methods based on deep learning. This comparison focuses on how well methods generalize rather than a quantitative comparison on a specific dataset. This study shows that while monocular depth estimation methods work well on images similar to training images, they often show artifacts when applied on images out of the training distribution. We evaluate the different methods with images similar to training data and images with unusual point of views (e.g. top-down) or paintings. The readers are invited to judge by themselves about the advantages and drawbacks of all methods by submitting their own images to the online demo associated with the present paper. Source Code The source codes and documentation for the algorithms presented in this paper are available from the web page of this article(1). Usage instructions are included in the README.md file of each archive. The original implementations of the methods are available at the following links: MiDaS and DPT methods(2), Adabins method(3), GLPDepth method(4), 3DShape method(5). This is an MLBriefs article, the source codes have not been reviewed!