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

Depth information is the foundation of perception, essential for autonomousdriving, robotics, and other source-constrained applications. Promptlyobtaining accurate and efficient depth information allows for a rapid responsein dynamic environments. Sensor-based methods using LIDAR and RADAR obtain highprecision at the cost of high power consumption, price, and volume. While dueto advances in deep learning, vision-based approaches have recently receivedmuch attention and can overcome these drawbacks. In this work, we explore anextreme scenario in vision-based settings: estimate a depth map from onemonocular image severely plagued by grid artifacts and blurry edges. To addressthis scenario, We first design a convolutional attention mechanism block (CAMB)which consists of channel attention and spatial attention sequentially andinsert these CAMBs into skip connections. As a result, our novel approach canfind the focus of current image with minimal overhead and avoid losses of depthfeatures. Next, by combining the depth value, the gradients of X axis, Y axisand diagonal directions, and the structural similarity index measure (SSIM), wepropose our novel loss function. Moreover, we utilize pixel blocks toaccelerate the computation of the loss function. Finally, we show, throughcomprehensive experiments on two large-scale image datasets, i.e. KITTI andNYU-V2, that our method outperforms several representative baselines.

Depth Monocular Estimation with Attention-based Encoder-Decoder Network from Single Image