Monocular Depth Estimation Using Neural Regression Forest

By Anirban Roy, Sinisa Todorovic. 2016

327 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Media Technology

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 Media Technology

Citations in the corpus

Black-Box vs. White-Box: Understanding Their Advantages and Weaknesses From a Practical Point of View (2019, background): “with the representation learning functionality known from deep convolutional networks by training them in an endto-end way [150]–[155].”
Towards Robust Monocular Depth Estimation: Mixing Datasets for Zero-Shot Cross-Dataset Transfer (2019, background): “vations have been proposed to enhance prediction accuracy [16], [17], [18], [19], [20].”
Deeper Depth Prediction with Fully Convolutional Residual Networks (2016, methodology): “CNN approaches have often been combined with CRF-based regularization, either as a post-processing step [16, 37] or via structured deep learning [19], as well as with random forests [27].”
NBDT: Neural-Backed Decision Trees (2020, background): “ 46, 35, 31]. However, these models forsake interpretability by creating impure leaves. Other approaches fuse deep learning into each decision tree node: an entire neural network [33], several layers [31, 38], a linear layer [1], or some other parameterization of neural network output [23]. These mo”
GeoNet: Geometric Neural Network for Joint Depth and Surface Normal Estimation (2018, background): “There exist a large amount of methods on depth estimation [25, 19, 8, 7, 21, 31, 24, 16, 20, 34, 18] and surface normal estimation [7, 33, 3, 2, 18] from a single image.”
Depth-Aware Video Frame Interpolation (2019, background): “Recently, several learning-based approaches [8, 9, 11, 18, 20, 31, 32, 37] aim to estimate the depth from a single image.”
Deep Depth Completion of a Single RGB-D Image (2018, background): “combined shallow convolutional networks with regression forests to reduce the need for large training sets [59].”
Multi-scale Continuous CRFs as Sequential Deep Networks for Monocular Depth Estimation (2017, background): “More recent approaches for depth estimation are based on CNNs [8, 20, 32, 26, 18].”
DeepLiDAR: Deep Surface Normal Guided Depth Prediction for Outdoor Scene From Sparse LiDAR Data and Single Color Image (2018, methodology): “With the development of deep learning, many methods [9, 25, 40, 28] based on deep neural networks have been proposed for the single-view depth estimation due to the strong feature representation of deep networks.”
MegaDepth: Learning Single-View Depth Prediction from Internet Photos (2018, background): “A standard approach is to collect RGB images with ground truth depth, and then train a model (e.g., a CNN) to predict depth from RGB [7, 22, 23, 27, 3, 19].”
Monocular 3D Human Pose Estimation by Predicting Depth on Joints (2017, background): “Predicting depth from monocular image is a long-standing problem and recent approaches [8, 35, 16, 5, 25] using deep neural networks have made a great progress in this area.”
Pattern-Affinitive Propagation Across Depth, Surface Normal and Semantic Segmentation (2019, background): “Depth Estimation: Many works have been proposed for monocular depth estimation [10, 11, 37, 32, 42, 29, 63, 54, 47, 60, 58, 46, 62].”
UM-Adapt: Unsupervised Multi-Task Adaptation Using Adversarial Cross-Task Distillation (2019, unknown): “[47] 795 0.”
PAD-Net: Multi-tasks Guided Prediction-and-Distillation Network for Simultaneous Depth Estimation and Scene Parsing (2018, methodology): “The second group of the methods is based on the advanced deep learning models [9, 32, 51, 43, 28].”
Sparse-to-Dense: Depth Prediction from Sparse Depth Samples and a Single Image (2017, methodology): “1) NYU-Depth-v2 Dataset: We compare with RGB-based approaches [3, 13, 30], as well as the fusion approach [24] that utilizes an additional 2D laser scanner mounted on a ground robot.”
Deep Ordinal Regression Network for Monocular Depth Estimation (2018, methodology): “To improve efﬁciency, Roy and Todorovic [48] proposed the Neural Regression Forest method which allows for parallelizable training of “shallow” CNNs. Recently, unsupervised or semi-supervised learning is introduced to learn depth estimation networks [17, 33].”
Monocular depth estimation based on deep learning: An overview (2020, background): “2] 2015 RGB + Depth p Fully CNNs Eigen et al. [51] 2015 RGB + Depth p Multi-task Szegedy et al. [59] 2015 RGB + Depth p Inception Module Mousavian et al. [71] 2016 RGB + Depth p Multi-task Roy et al. [73] 2016 RGB + Depth p RFs Mayer et al. [33] 2016 RGB + Disparity p Multi-task Laina et al. [54] 20”