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 Plus annual

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

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Self-supervised monocular depth estimation that does not require ground truthfor training has attracted attention in recent years. It is of high interest todesign lightweight but effective models so that they can be deployed on edgedevices. Many existing architectures benefit from using heavier backbones atthe expense of model sizes. This paper achieves comparable results with alightweight architecture. Specifically, the efficient combination of CNNs andTransformers is investigated, and a hybrid architecture called Lite-Mono ispresented. A Consecutive Dilated Convolutions (CDC) module and a Local-GlobalFeatures Interaction (LGFI) module are proposed. The former is used to extractrich multi-scale local features, and the latter takes advantage of theself-attention mechanism to encode long-range global information into thefeatures. Experiments demonstrate that Lite-Mono outperforms Monodepth2 by alarge margin in accuracy, with about 80% fewer trainable parameters.

Lite-Mono: A Lightweight CNN and Transformer Architecture for Self-Supervised Monocular Depth Estimation