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

In self-supervised monocular depth estimation, the depth discontinuity andmotion objects' artifacts are still challenging problems. Existingself-supervised methods usually utilize a single view to train the depthestimation network. Compared with static views, abundant dynamic propertiesbetween video frames are beneficial to refined depth estimation, especially fordynamic objects. In this work, we propose a novel self-supervised jointlearning framework for depth estimation using consecutive frames from monocularand stereo videos. The main idea is using an implicit depth cue extractor whichleverages dynamic and static cues to generate useful depth proposals. Thesecues can predict distinguishable motion contours and geometric scenestructures. Furthermore, a new high-dimensional attention module is introducedto extract clear global transformation, which effectively suppressesuncertainty of local descriptors in high-dimensional space, resulting in a morereliable optimization in learning framework. Experiments demonstrate that theproposed framework outperforms the state-of-the-art(SOTA) on KITTI and Make3Ddatasets.

Self-Supervised Joint Learning Framework of Depth Estimation via Implicit Cues