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

Given two consecutive frames, video interpolation aims at generatingintermediate frame(s) to form both spatially and temporally coherent videosequences. While most existing methods focus on single-frame interpolation, wepropose an end-to-end convolutional neural network for variable-lengthmulti-frame video interpolation, where the motion interpretation and occlusionreasoning are jointly modeled. We start by computing bi-directional opticalflow between the input images using a U-Net architecture. These flows are thenlinearly combined at each time step to approximate the intermediatebi-directional optical flows. These approximate flows, however, only work wellin locally smooth regions and produce artifacts around motion boundaries. Toaddress this shortcoming, we employ another U-Net to refine the approximatedflow and also predict soft visibility maps. Finally, the two input images arewarped and linearly fused to form each intermediate frame. By applying thevisibility maps to the warped images before fusion, we exclude the contributionof occluded pixels to the interpolated intermediate frame to avoid artifacts.Since none of our learned network parameters are time-dependent, our approachis able to produce as many intermediate frames as needed. We use 1,132 videoclips with 240-fps, containing 300K individual video frames, to train ournetwork. Experimental results on several datasets, predicting different numbersof interpolated frames, demonstrate that our approach performs consistentlybetter than existing methods.

Super SloMo: High Quality Estimation of Multiple Intermediate Frames for Video Interpolation