Skip to Main content Skip to Navigation
Conference papers

Deep Ordinal Regression Network for Monocular Depth Estimation

Abstract : Monocular depth estimation, which plays a crucial role in understanding 3D scene geometry, is an ill-posed problem. Recent methods have gained significant improvement by exploring image-level information and hierarchical features from deep convolutional neural networks (DCNNs). These methods model depth estimation as a regression problem and train the regression networks by minimizing mean squared error, which suffers from slow convergence and unsatisfactory local solutions. Besides, existing depth estimation networks employ repeated spatial pooling operations, resulting in undesirable low-resolution feature maps. To obtain high-resolution depth maps, skip-connections or multi-layer deconvolution networks are required, which complicates network training and consumes much more computations. To eliminate or at least largely reduce these problems, we introduce a spacing-increasing discretization (SID) strategy to discretize depth and recast depth network learning as an ordinal regression problem. By training the network using an ordinary regression loss, our method achieves much higher accuracy and faster convergence in synch. Furthermore, we adopt a multi-scale network structure which avoids unnecessary spatial pooling and captures multi-scale information in parallel. The proposed deep ordinal regression network (DORN) achieves state-of-the-art results on three challenging benchmarks, i.e., KITTI [16], Make3D [49], and NYU Depth v2 [41], and outperforms existing methods by a large margin.
Document type :
Conference papers
Complete list of metadata

Cited literature [62 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-01741163
Contributor : Chaohui Wang Connect in order to contact the contributor
Submitted on : Thursday, March 22, 2018 - 5:18:37 PM
Last modification on : Tuesday, October 19, 2021 - 11:26:22 AM
Long-term archiving on: : Thursday, September 13, 2018 - 10:42:03 AM

File

CVPR18_DepthEstimation.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01741163, version 1

Citation

Huan Fu, Mingming Gong, Chaohui Wang, Kayhan Batmanghelich, Dacheng Tao. Deep Ordinal Regression Network for Monocular Depth Estimation. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Jun 2018, Salt Lake City, United States. ⟨hal-01741163⟩

Share

Metrics

Record views

1152

Files downloads

4693