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

How to efficiently utilize temporal information to recover videos in aconsistent way is the main issue for video inpainting problems. Conventional 2DCNNs have achieved good performance on image inpainting but often lead totemporally inconsistent results where frames will flicker when applied tovideos (seehttps://www.youtube.com/watch?v=87Vh1HDBjD0&list=PLPoVtv-xp_dL5uckIzz1PKwNjg1yI0I94&index=1);3D CNNs can capture temporal information but are computationally intensive andhard to train. In this paper, we present a novel component termed LearnableGated Temporal Shift Module (LGTSM) for video inpainting models that couldeffectively tackle arbitrary video masks without additional parameters from 3Dconvolutions. LGTSM is designed to let 2D convolutions make use of neighboringframes more efficiently, which is crucial for video inpainting. Specifically,in each layer, LGTSM learns to shift some channels to its temporal neighbors sothat 2D convolutions could be enhanced to handle temporal information.Meanwhile, a gated convolution is applied to the layer to identify the maskedareas that are poisoning for conventional convolutions. On the FaceForensicsand Free-form Video Inpainting (FVI) dataset, our model achievesstate-of-the-art results with simply 33% of parameters and inference time.

Learnable Gated Temporal Shift Module for Deep Video Inpainting