Self-Supervised Monocular Depth Estimation With Multiscale Perception

Zhang, Yourun; Gong, Maoguo; Li, Jianzhao; Zhang, Mingyang; Jiang, Fenlong; Zhao, Hongyu

doi:10.1109/tip.2022.3167307

Cited by 32 publications

(15 citation statements)

References 48 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To get sharper boundaries in depth maps, Song et al [21] combine structure attention and global attention to intensify the extracted features locally and globally. Zhang et al [22] propose a multi-scale strategy to extend the perceptual area for photometric error, which leads to more accurate depth estimation results than previous methods.…”

Section: Depth Predictionmentioning

confidence: 99%

See 1 more Smart Citation

Self‐supervised depth completion with multi‐view geometric constraints

Xiong

Zhang

Liu

et al. 2023

IET Image Processing

View full text Add to dashboard Cite

Self‐supervised learning‐based depth completion is a cost‐effective way for 3D environment perception. However, it is also a challenging task because sparse depth may deactivate neural networks. In this paper, a novel Sparse‐Dense Depth Consistency Loss (SDDCL) is proposed to penalize not only the estimated depth map with sparse input points but also consecutive completed dense depth maps. Combined with the pose consistency loss, a new self‐supervised learning scheme is developed, using multi‐view geometric constraints, to achieve more accurate depth completion results. Moreover, to tackle the sparsity issue of input depth, a Quasi Dense Representations (QDR) module with triplet branches for spatial pyramid pooling is proposed to produce more dense feature maps. Extensive experimental results on VOID, NYUv2, and KITTI datasets show that the method outperforms state‐of‐the‐art self‐supervised depth completion methods.

show abstract

Section: Depth Predictionmentioning

confidence: 99%

“…Zhang et al. [22] propose a multi‐scale strategy to extend the perceptual area for photometric error, which leads to more accurate depth estimation results than previous methods.…”

Section: Related Workmentioning

confidence: 99%

Self‐supervised depth completion with multi‐view geometric constraints

Xiong

Zhang

Liu

et al. 2023

IET Image Processing

View full text Add to dashboard Cite

show abstract

“…To overcome the above limitation, many multi-scale CNN have further promoted computer vision [40]. For instance, Zhao et al designed a pyramid scene parsing network (PSPNet) for semantic segmentation [41].…”

Section: Related Workmentioning

confidence: 99%

Frequency Spectrum Intensity Attention Network for Building Detection from High-Resolution Imagery

2022

View full text Add to dashboard Cite

Computational intelligence techniques have been widely used for automatic building detection from high-resolution remote sensing imagery and especially the methods based on neural networks. However, existing methods do not pay attention to the value of high-frequency and low-frequency information in the frequency domain for feature extraction of buildings in remote sensing images. To overcome these limitations, this paper proposes a frequency spectrum intensity attention network (FSIANet) with an encoder–decoder structure for automatic building detection. The proposed FSIANet mainly involves two innovations. One, a novel and plug-and-play frequency spectrum intensity attention (FSIA) mechanism is devised to enhance feature representation by evaluating the informative abundance of the feature maps. The FSIA is deployed after each convolutional block in the proposed FSIANet. Two, an atrous frequency spectrum attention pyramid (AFSAP) is constructed by introducing FSIA in widely used atrous spatial pyramid pooling. The AFSAP is able to select the features with high response to building semantic features at each scale and weaken the features with low response, thus enhancing the feature representation of buildings. The proposed FSIANet is evaluated on two large public datasets (East Asia and Inria Aerial Image Dataset), which demonstrates that the proposed method can achieve the state-of-the-art performance in terms of F1-score and intersection-over-union.

show abstract

“…Zhang et al. [ 13,14 ] addressed the problem of recovering object depth from images, analyzed the detailed features of a single image, and combined with spatial pyramid pooling (SPP), constructed a convolutional spatial propagation network (CSPN) model to obtain accurate global dense depth maps. Nikoohemat et al.…”

Section: Introductionmentioning

confidence: 99%

“…The 3D reconstruction of large-scale outdoor scenes, especially efficient and robust reconstruction, has attracted more and more attention in both civil and military fields. Zhang et al [13,14] addressed the problem of recovering object depth from images, analyzed the detailed features of a single image, and combined with spatial pyramid pooling (SPP), constructed a convolutional spatial propagation network (CSPN) model to obtain accurate global dense depth maps. Nikoohemat et al [15,16] aimed at the high demand in the architecture industry for existing building point cloud scanning (scan-to-BIM), combined with the complexity of building structures, and the variety of environmental objects to construct an automatic reconstruction of a BIM wall, its topological method model to obtain point cloud diagrams of different wall axes and connection types, and can process multilevel structures simultaneously.…”

Section: Introductionmentioning

confidence: 99%

A Coupled 3d Morphological Reconstruction Approach for Surface Microcrack in Si₃n₄ Ceramic Bearing Roller Based on Adaptive Nano Feature Extraction & Multiscale Depth Fusion

Liao

et al. 2023

Small Methods

View full text Add to dashboard Cite

To extract the fuzzy contour features, tiny depth features of surface microcracks in the Si3N4 ceramic bearings roller. An adaptive nano feature extraction and multiscale deep fusion coupling method is proposed, to sufficiently reconstruct the three‐dimensional morphology characteristics of surface microcracks. Construct an adaptive nano feature extraction method, form the surface microcrack image scale space and the Gaussian difference pyramid function equation, realize the detection and matching of global feature points. The sparse point cloud is obtained. Through polar‐line correction, depth estimation, and fusion of feature points on the surface microcracks image, a multiscale depth fusion matching cost pixel function is established to realize a dense point cloud reconstruction of surface microcracks. The reconstruction results show that the highest value of the local convex surface reconstructed by the dense point cloud reaches 1183 nm, and the lowest local concave surface is accurate to 296 nm. Compared with the measurement results of the confocal platform, the relative error of the reconstruction result is 24.6%. The overall feature‐matching rate of the reconstruction reaches 93.3%. It provides a theoretical basis for the study of surface microcrack propagation mechanism and the prediction of bearing life.

show abstract

Self-Supervised Monocular Depth Estimation With Multiscale Perception

Cited by 32 publications

References 48 publications

Self‐supervised depth completion with multi‐view geometric constraints

Self‐supervised depth completion with multi‐view geometric constraints

Frequency Spectrum Intensity Attention Network for Building Detection from High-Resolution Imagery

A Coupled 3d Morphological Reconstruction Approach for Surface Microcrack in Si₃n₄ Ceramic Bearing Roller Based on Adaptive Nano Feature Extraction & Multiscale Depth Fusion

Contact Info

Product

Resources

About

Self-Supervised Monocular Depth Estimation With Multiscale Perception

Cited by 32 publications

References 48 publications

Self‐supervised depth completion with multi‐view geometric constraints

Self‐supervised depth completion with multi‐view geometric constraints

Frequency Spectrum Intensity Attention Network for Building Detection from High-Resolution Imagery

A Coupled 3d Morphological Reconstruction Approach for Surface Microcrack in Si3n4 Ceramic Bearing Roller Based on Adaptive Nano Feature Extraction & Multiscale Depth Fusion

Contact Info

Product

Resources

About

A Coupled 3d Morphological Reconstruction Approach for Surface Microcrack in Si₃n₄ Ceramic Bearing Roller Based on Adaptive Nano Feature Extraction & Multiscale Depth Fusion