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

This work aims to estimate a high-quality depth map from a single RGB image.Due to the lack of depth clues, making full use of the long-range correlationand the local information is critical for accurate depth estimation. Towardsthis end, we introduce an uncertainty rectified cross-distillation betweenTransformer and convolutional neural network (CNN) to learn a unified depthestimator. Specifically, we use the depth estimates from the Transformer branchand the CNN branch as pseudo labels to teach each other. Meanwhile, we modelthe pixel-wise depth uncertainty to rectify the loss weights of noisy pseudolabels. To avoid the large capacity gap induced by the strong Transformerbranch deteriorating the cross-distillation, we transfer the feature maps fromTransformer to CNN and design coupling units to assist the weak CNN branch toleverage the transferred features. Furthermore, we propose a surprisinglysimple yet highly effective data augmentation technique CutFlip, which enforcesthe model to exploit more valuable clues apart from the vertical image positionfor depth inference. Extensive experiments demonstrate that our model,termed~\textbf{URCDC-Depth}, exceeds previous state-of-the-art methods on theKITTI, NYU-Depth-v2 and SUN RGB-D datasets, even with no additionalcomputational burden at inference time. The source code is publicly availableat \url{https://github.com/ShuweiShao/URCDC-Depth}.

URCDC-Depth: Uncertainty Rectified Cross-Distillation with CutFlip for Monocular Depth Estimation