Going Deep: An Introduction to Depth Estimation with Fully Convolutional Residual Networks

Gif by author

Have you ever looked at a two-dimensional image and wished you could know the depth of the objects in the scene? Perhaps you’re a computer vision researcher trying to build an autonomous vehicle that can accurately gauge distances, or a filmmaker seeking to create a more immersive virtual reality experience. Whatever your motivations may be, depth estimation is a fascinating and challenging task that has applications in a wide range of fields.

In this blog post, we’ll explore the task of depth estimation and its use cases. We’ll then dive into the details of how fully convolutional residual networks work, and show how they can achieve state-of-the-art results in depth estimation. Whether you’re a seasoned computer vision expert or just getting started in the field, you’re sure to learn something new about this exciting area of research.

Introduction & use cases

Depth estimation involves determining the distances between objects in a scene and the viewer’s point of view. Traditionally, this has been done with specialized hardware such as stereo cameras or depth sensors, but recent advancements in deep learning have led to the development of fully convolutional residual networks (FCRN) that can estimate depth from 2D images…