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

We consider the problem of temporal view synthesis, where the goal is topredict a future video frame from the past frames using knowledge of the depthand relative camera motion. In contrast to revealing the disoccluded regionsthrough intensity based infilling, we study the idea of an infilling vector toinfill by pointing to a non-disoccluded region in the synthesized view. Toexploit the structure of disocclusions created by camera motion during theirinfilling, we rely on two important cues, temporal correlation of infillingdirections and depth. We design a learning framework to predict the infillingvector by computing a temporal prior that reflects past infilling directionsand a normalized depth map as input to the network. We conduct extensiveexperiments on a large scale dataset we build for evaluating temporal viewsynthesis in addition to the SceneNet RGB-D dataset. Our experimentsdemonstrate that our infilling vector prediction approach achieves superiorquantitative and qualitative infilling performance compared to other approachesin literature.

Revealing Disocclusions in Temporal View Synthesis through Infilling Vector Prediction