Single-Image Depth Perception in the Wild

Abstract

This paper studies single-image depth perception in the wild, i.e., recovering depth from a single image taken in unconstrained settings. We introduce a new dataset "Depth in the Wild" consisting of images in the wild annotated with relative depth between pairs of random points. We also propose a new algorithm that learns to estimate metric depth using annotations of relative depth. Compared to the state of the art, our algorithm is simpler and performs better. Experiments show that our algorithm, combined with existing RGB-D data and our new relative depth annotations, significantly improves single-image depth perception in the wild.

References in corpus (8)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Depth Map Prediction from a Single Image using a Multi-Scale Deep Network
• Stacked Hourglass Networks for Human Pose Estimation
• DepthTransfer: Depth Extraction from Video Using Non-parametric Sampling
• Holistically-Nested Edge Detection
• A Large Dataset of Object Scans
• Monocular Object Instance Segmentation and Depth Ordering with CNNs
• Coupled Depth Learning

Cited by in corpus (33)

• Monocular Depth Estimation Based On Deep Learning: An Overview
• Estimating Depth from RGB and Sparse Sensing
• Semi-Supervised Deep Learning for Monocular Depth Map Prediction
• MegaDepth: Learning Single-View Depth Prediction from Internet Photos
• Learning Feature Pyramids for Human Pose Estimation
• Unsupervised Monocular Depth Estimation with Left-Right Consistency
• Depth-aware Blending of Smoothed Images for Bokeh Effect Generation
• Deep Depth Completion of a Single RGB-D Image
• Unsupervised Learning of Monocular Depth Estimation with Bundle Adjustment, Super-Resolution and Clip Loss
• Revisiting Single Image Depth Estimation: Toward Higher Resolution Maps with Accurate Object Boundaries
• Deep attention-based classification network for robust depth prediction
• Single View Stereo Matching
• Fast and Accurate Single-Image Depth Estimation on Mobile Devices, Mobile AI 2021 Challenge: Report
• Learning Shape Priors for Single-View 3D Completion and Reconstruction
• Semi-parametric Image Synthesis
• DRPose3D: Depth Ranking in 3D Human Pose Estimation
• InverseRenderNet: Learning single image inverse rendering
• Monocular Depth Estimation with Hierarchical Fusion of Dilated CNNs and Soft-Weighted-Sum Inference
• DF-Net: Unsupervised Joint Learning of Depth and Flow using Cross-Task Consistency
• Ordinal Depth Supervision for 3D Human Pose Estimation
• SIGNet: Semantic Instance Aided Unsupervised 3D Geometry Perception
• IMU-Assisted Learning of Single-View Rolling Shutter Correction
• Single Image Reflection Removal with Physically-Based Training Images
• DAN: A Deformation-Aware Network for Consecutive Biomedical Image Interpolation
• Self-Supervised Monocular Image Depth Learning and Confidence Estimation
• Region Deformer Networks for Unsupervised Depth Estimation from Unconstrained Monocular Videos
• Relative Depth Order Estimation Using Multi-scale Densely Connected Convolutional Networks
• Reflectance Adaptive Filtering Improves Intrinsic Image Estimation
• Unsupervised monocular stereo matching
• Depth Assisted Full Resolution Network for Single Image-based View Synthesis
• Deep Depth from Defocus: how can defocus blur improve 3D estimation using dense neural networks?
• 3D Surface Reconstruction by Pointillism
• Soccer on Your Tabletop