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

In this paper, we address monocular depth estimation with deep neuralnetworks. To enable training of deep monocular estimation models with varioussources of datasets, state-of-the-art methods adopt image-level normalizationstrategies to generate affine-invariant depth representations. However,learning with image-level normalization mainly emphasizes the relations ofpixel representations with the global statistic in the images, such as thestructure of the scene, while the fine-grained depth difference may beoverlooked. In this paper, we propose a novel multi-scale depth normalizationmethod that hierarchically normalizes the depth representations based onspatial information and depth distributions. Compared with previousnormalization strategies applied only at the holistic image level, the proposedhierarchical normalization can effectively preserve the fine-grained detailsand improve accuracy. We present two strategies that define the hierarchicalnormalization contexts in the depth domain and the spatial domain,respectively. Our extensive experiments show that the proposed normalizationstrategy remarkably outperforms previous normalization methods, and we set newstate-of-the-art on five zero-shot transfer benchmark datasets.

Hierarchical Normalization for Robust Monocular Depth Estimation