Depth and surface normal estimation from monocular images using regression on deep features and hierarchical CRFs

By Bo Li, Chunhua Shen, Yuchao Dai, Anton van den Hengel, Mingyi He. 2015

642 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Industrial and Manufacturing Engineering

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Industrial and Manufacturing Engineering

Citations in the corpus

Matterport3D: Learning from RGB-D Data in Indoor Environments (2017, methodology): “Networks have been trained to perform that task us-ing RGB-D datasets in the past [9, 23, 3, 41, 48].”
Unsupervised CNN for Single View Depth Estimation: Geometry to the Rescue (2016, background): “Trained initially on tasks such as image classiﬁcation, and ﬁne-tuned to ﬁt other tasks, supervised CNNs are now state of the art for object detection [11], per-pixel image classiﬁcation [19], depth and normal prediction from single image [15], human pose estimation [7] and many other applications.”
Monocular depth estimation based on deep learning: An overview (2020, methodology): “[66] propose a refinement method based on the hierarchical conditional random fields (CRFs), which is also widely used for semantic segmentation [67], [68].”
Shading Annotations in the Wild (2017, background): “These datasets can be used to train algorithms to predict depth or surface normals from a single image [34, 11, 25, 2, 9].”
Metric3D v2: A Versatile Monocular Geometric Foundation Model for Zero-Shot Metric Depth and Surface Normal Estimation (2024, positive): “Currently, learning-based methods [32], [33], [34], [42], [64], [65], [66], [67], [68], [69], [70], [71], [72] have dominated monocular surface normal estimation.”
Pattern-Affinitive Propagation Across Depth, Surface Normal and Semantic Segmentation (2019, background): “Method data rmse rel log δ1 δ2 δ3 HCRF [32] 795 0.”
Depth Estimation via Affinity Learned with Convolutional Spatial Propagation Network (2018, background): “Others tried to improve the estimated details further by appending a conditional random field (CRF) [23,24,25] and joint training [26,27].”
Every Pixel Counts ++: Joint Learning of Geometry and Motion with 3D Holistic Understanding (2018, positive): “Others try to improve the estimation further by appending a conditional random field (CRF) [46], [47], [48], [49].”
Is Pseudo-Lidar needed for Monocular 3D Object detection? (2021, background): “It is itself a thriving research area: the community has pushed toward accurate dense depth prediction via supervised [35, 29, 19, 50, 26, 18, 32, 13, 31] as well as self-supervised [48, 75, 62, 16, 15, 64] methods.”
Deep Ordinal Regression Network for Monocular Depth Estimation (2018, neutral): “ 2 3 rel log 10 rms Make3D [51] 0.447 0.745 0.897 0.349 - 1.214 DepthTransfer [28] - - - 0.35 0.131 1.2 Liu et al. [41] - - - 0.335 0.127 1.06 Ladicky et al. [34] 0.542 0.829 0.941 - - - Li et al. [36] 0.621 0.886 0.968 0.232 0.094 0.821 Wang et al. [57] 0.605 0.890 0.970 0.220 - 0.824 Roy et al. [”
Single-Image Depth Perception in the Wild (2016, background): “neural networks and RGB-D datasets [4, 5] has led to major advances [27, 6, 28, 8, 10, 14].”
Monocular human pose estimation: A survey of deep learning-based methods (2020, methodology): “Model-free (1) Single-stage : direct prediction (Li and Chan, 2014; Pavlakos et al., 2017), body structure constraint (Li et al., 2015b; Tekin et al., 2016; Sun et al., 2017; Pavlakos et al., 2018a) (2) 2D-to-3D : (Martinez et al., 2017; Zhou et al., 2017; Tekin et al., 2017; Li and Lee, 2019;…”
Visual SLAM for Automated Driving: Exploring the Applications of Deep Learning (2018, neutral): “When the depth estimation is supervised, the loss function usually reads as regression loss [18, 39, 41, 59] w/wo regularization terms [70].”
DF-Net: Unsupervised Joint Learning of Depth and Flow using Cross-Task Consistency (2018, background): “Single-view depth model learning can also benefit from joint training with surface normal estimation [35,67] or semantic segmentation [13,30].”
DeepLiDAR: Deep Surface Normal Guided Depth Prediction for Outdoor Scene From Sparse LiDAR Data and Single Color Image (2018, methodology): “There are also some methods [8, 26, 4, 39] which exploit surface normals during the depth estimation.”