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

For visual estimation of optical flow, a crucial function for many visiontasks, unsupervised learning, using the supervision of view synthesis hasemerged as a promising alternative to supervised methods, since ground-truthflow is not readily available in many cases. However, unsupervised learning islikely to be unstable when pixel tracking is lost due to occlusion and motionblur, or the pixel matching is impaired due to variation in image content andspatial structure over time. In natural environments, dynamic occlusion orobject variation is a relatively slow temporal process spanning several frames.We, therefore, explore the optical flow estimation from multiple-framesequences of dynamic scenes, whereas most of the existing unsupervisedapproaches are based on temporal static models. We handle the unsupervisedoptical flow estimation with a temporal dynamic model by introducing aspatial-temporal dual recurrent block based on the predictive coding structure,which feeds the previous high-level motion prior to the current optical flowestimator. Assuming temporal smoothness of optical flow, we use motion priorsof the adjacent frames to provide more reliable supervision of the occludedregions. To grasp the essence of challenging scenes, we simulate variousscenarios across long sequences, including dynamic occlusion, contentvariation, and spatial variation, and adopt self-supervised distillation tomake the model understand the object's motion patterns in a prolonged dynamicenvironment. Experiments on KITTI 2012, KITTI 2015, Sintel Clean, and SintelFinal datasets demonstrate the effectiveness of our methods on unsupervisedoptical flow estimation. The proposal achieves state-of-the-art performancewith advantages in memory overhead.

Unsupervised Learning Optical Flow in Multi-frame Dynamic Environment Using Temporal Dynamic Modeling