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

Monocular depth estimators either require explicit scale supervision throughauxiliary sensors or suffer from scale ambiguity, which renders them difficultto deploy in downstream applications. A possible source of scale is the sizesof objects found in the scene, but inaccurate localization makes them difficultto exploit. In this paper, we introduce a novel scale-aware monocular depthestimation method called StableCamH that does not require any auxiliary sensoror supervision. The key idea is to exploit prior knowledge of object heights inthe scene but aggregate the height cues into a single invariant measure commonto all frames in a road video sequence, namely the camera height. Byformulating monocular depth estimation as camera height optimization, weachieve robust and accurate unsupervised end-to-end training. To realizeStableCamH, we devise a novel learning-based size prior that can directlyconvert car appearance into its dimensions. Extensive experiments on KITTI andCityscapes show the effectiveness of StableCamH, its state-of-the-art accuracycompared with related methods, and its generalizability. The training frameworkof StableCamH can be used for any monocular depth estimation method and willhopefully become a fundamental building block for further work.

Camera Height Doesn't Change: Unsupervised Monocular Scale-Aware Road-Scene Depth Estimation