English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Zendy Plus

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Zendy Tools

Zendy Open

Self-supervised depth learning from monocular images normally relies on the2D pixel-wise photometric relation between temporally adjacent image frames.However, they neither fully exploit the 3D point-wise geometriccorrespondences, nor effectively tackle the ambiguities in the photometricwarping caused by occlusions or illumination inconsistency. To address theseproblems, this work proposes Density Volume Construction Network (DevNet), anovel self-supervised monocular depth learning framework, that can consider 3Dspatial information, and exploit stronger geometric constraints among adjacentcamera frustums. Instead of directly regressing the pixel value from a singleimage, our DevNet divides the camera frustum into multiple parallel planes andpredicts the pointwise occlusion probability density on each plane. The finaldepth map is generated by integrating the density along corresponding rays.During the training process, novel regularization strategies and loss functionsare introduced to mitigate photometric ambiguities and overfitting. Withoutobviously enlarging model parameters size or running time, DevNet outperformsseveral representative baselines on both the KITTI-2015 outdoor dataset andNYU-V2 indoor dataset. In particular, the root-mean-square-deviation is reducedby around 4% with DevNet on both KITTI-2015 and NYU-V2 in the task of depthestimation. Code is available at https://github.com/gitkaichenzhou/DevNet.

DevNet: Self-supervised Monocular Depth Learning via Density Volume Construction