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/

Global: Zendy Tools annual

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

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Self-supervised depth estimation has made a great success in learning depthfrom unlabeled image sequences. While the mappings between image and pixel-wisedepth are well-studied in current methods, the correlation between image, depthand scene semantics, however, is less considered. This hinders the network tobetter understand the real geometry of the scene, since the contextual clues,contribute not only the latent representations of scene depth, but also thestraight constraints for depth map. In this paper, we leverage the two benefitsby proposing the implicit and explicit semantic guidance for accurateself-supervised depth estimation. We propose a Semantic-aware Spatial FeatureAlignment (SSFA) scheme to effectively align implicit semantic features withdepth features for scene-aware depth estimation. We also propose asemantic-guided ranking loss to explicitly constrain the estimated depth mapsto be consistent with real scene contextual properties. Both semantic labelnoise and prediction uncertainty is considered to yield reliable depthsupervisions. Extensive experimental results show that our method produces highquality depth maps which are consistently superior either on complex scenes ordiverse semantic categories, and outperforms the state-of-the-art methods by asignificant margin.

Learning Depth via Leveraging Semantics: Self-supervised Monocular Depth Estimation with Both Implicit and Explicit Semantic Guidance