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

Deep convolutional neural networks (DCNN) have recently shown promisingresults in low-level computer vision problems such as optical flow anddisparity estimation, but still, have much room to further improve theirperformance. In this paper, we propose a novel sub-pixel convolution-basedencoder-decoder network for optical flow and disparity estimations, which canextend FlowNetS and DispNet by replacing the deconvolution layers withsup-pixel convolution blocks. By using sub-pixel refinement and estimation onthe decoder stages instead of deconvolution, we can significantly improve theestimation accuracy for optical flow and disparity, even with reduced numbersof parameters. We show a supervised end-to-end training of our proposednetworks for optical flow and disparity estimations, and an unsupervisedend-to-end training for monocular depth and pose estimations. In order toverify the effectiveness of our proposed networks, we perform intensiveexperiments for (i) optical flow and disparity estimations, and (ii) monoculardepth and pose estimations. Throughout the extensive experiments, our proposednetworks outperform the baselines such as FlowNetS and DispNet in terms ofestimation accuracy and training times.

Finding Correspondences for Optical Flow and Disparity Estimations using a Sub-pixel Convolution-based Encoder-Decoder Network