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

Self-supervised monocular depth estimation, aiming to learn scene depths fromsingle images in a self-supervised manner, has received much attentionrecently. In spite of recent efforts in this field, how to learn accurate scenedepths and alleviate the negative influence of occlusions for self-superviseddepth estimation, still remains an open problem. Addressing this problem, wefirstly empirically analyze the effects of both the continuous and discretedepth constraints which are widely used in the training process of manyexisting works. Then inspired by the above empirical analysis, we propose anovel network to learn an Occlusion-aware Coarse-to-Fine Depth map forself-supervised monocular depth estimation, called OCFD-Net. Given an arbitrarytraining set of stereo image pairs, the proposed OCFD-Net does not only employa discrete depth constraint for learning a coarse-level depth map, but alsoemploy a continuous depth constraint for learning a scene depth residual,resulting in a fine-level depth map. In addition, an occlusion-aware module isdesigned under the proposed OCFD-Net, which is able to improve the capabilityof the learnt fine-level depth map for handling occlusions. Experimentalresults on KITTI demonstrate that the proposed method outperforms thecomparative state-of-the-art methods under seven commonly used metrics in mostcases. In addition, experimental results on Make3D demonstrate theeffectiveness of the proposed method in terms of the cross-datasetgeneralization ability under four commonly used metrics. The code is availableat https://github.com/ZM-Zhou/OCFD-Net_pytorch.

Learning Occlusion-Aware Coarse-to-Fine Depth Map for Self-supervised Monocular Depth Estimation