Monocular Depth Estimation: A Survey

Abstract

Monocular depth estimation is often described as an ill-posed and inherently ambiguous problem. Estimating depth from 2D images is a crucial step in scene reconstruction, 3Dobject recognition, segmentation, and detection. The problem can be framed as: given a single RGB image as input, predict a dense depth map for each pixel. This problem is worsened by the fact that most scenes have large texture and structural variations, object occlusions, and rich geometric detailing. All these factors contribute to difficulty in accurate depth estimation. In this paper, we review five papers that attempt to solve the depth estimation problem with various techniques including supervised, weakly-supervised, and unsupervised learning techniques. We then compare these papers and understand the improvements made over one another. Finally, we explore potential improvements that can aid to better solve this problem.

8 pages, 1 figure, 4 tables

References in corpus (2)

• Depth Map Prediction from a Single Image using a Multi-Scale Deep Network
• Deep Ordinal Regression Network for Monocular Depth Estimation

Cited by in corpus (19)

• 3D Scene Geometry Estimation from 360${}^{\circ }$ Imagery: A Survey
• Learning the Pareto Front with Hypernetworks
• Detecting and Identifying Optical Signal Attacks on Autonomous Driving Systems
• Holopix50k: A Large-Scale In-the-wild Stereo Image Dataset
• Leveraging a realistic synthetic database to learn Shape-from-Shading for estimating the colon depth in colonoscopy images
• MonSter: Awakening the Mono in Stereo
• Towards Real-Time Monocular Depth Estimation for Robotics: A Survey
• RGB-D-Fusion: Image Conditioned Depth Diffusion of Humanoid Subjects
• Black-box Adversarial Attacks on Monocular Depth Estimation Using Evolutionary Multi-objective Optimization
• Progressive Depth Learning for Single Image Dehazing
• Pix2Surf: Learning Parametric 3D Surface Models of Objects from Images
• Monocular Spherical Depth Estimation with Explicitly Connected Weak Layout Cues
• Marine vessel tracking using a monocular camera
• Low-viewpoint forest depth dataset for sparse rover swarms
• Close-Proximity Underwater Terrain Mapping Using Learning-based Coarse Range Estimation
• A Dark Flash Normal Camera
• A Robust Billboard-based Free-viewpoint Video Synthesizing Algorithm for Sports Scenes
• Unpaired Single-Image Depth Synthesis with cycle-consistent Wasserstein GANs
• Improved Point Transformation Methods For Self-Supervised Depth Prediction