Underwater self-supervised monocular depth estimation and its application in image enhancement

Wang, Junting; Ye, Xiufen; Liu, Yusong; Mei, Xinkui; Hou, Jun

doi:10.1016/j.engappai.2023.105846

Cited by 13 publications

(2 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The gradient loss function is powerful on continuous smooth surfaces. 4 The x-gradient of the predicted depth map ∇ x (pred) and the x-gradient of the target depth map ∇ x (target) are calculated. Meanwhile, The y-gradient of the predicted depth map ∇ y (pred) and the y-gradient of the target depth map ∇ y (target) are calculated:…”

Section: Edge Gradient Lossmentioning

confidence: 99%

“…1,3 Currently, monocular endoscopic depth estimation is mainly self-supervised. [2][3][4] The method 2 focuses on generating parallax maps through methods such as SfM, which are then transformed into information such as depth maps or feature points. This information is projected back to the video frame as a self-supervised signal, which allows for the computation of the loss function.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Asymmetric edge-aware transformers for monocular endoscopic depth estimation

Wu,

Qi,

Fan

et al. 2024

Medical Imaging 2024: Image-Guided Procedures, Robotic Interventions, and Modeling

View full text Add to dashboard Cite

Monocular depth estimation is a popular task. Due to the difficulty of obtaining true depth labels for the bronchus and the characteristics of the bronchial image such as scarcity of texture, smoother surfaces and more holes, there are many challenges in bronchial depth estimation. Hence, we propose to use a ray tracing algorithm to generate virtual images along with their corresponding depth maps to train an asymmetric encoder-decoder transformer network for bronchial depth estimation. We propose the edge-aware unit to enhance the awareness of the bronchial internal structure considering that the bronchus has few texture features and many edges and holes. And asymmetric encoder-decoder is proposed by us for multi-layer features fusion. The experimental results of the virtual bronchial demonstrate that our method achieves the best results in several metrics, including MAE of 0.915 ± 0.596 and RMSE of 1.471 ± 1.097.

show abstract

Section: Edge Gradient Lossmentioning

confidence: 99%