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

Unsupervised optical flow estimation is especially hard near occlusions andmotion boundaries and in low-texture regions. We show that additionalinformation such as semantics and domain knowledge can help better constrainthis problem. We introduce SemARFlow, an unsupervised optical flow networkdesigned for autonomous driving data that takes estimated semantic segmentationmasks as additional inputs. This additional information is injected into theencoder and into a learned upsampler that refines the flow output. In addition,a simple yet effective semantic augmentation module provides self-supervisionwhen learning flow and its boundaries for vehicles, poles, and sky. Together,these injections of semantic information improve the KITTI-2015 optical flowtest error rate from 11.80% to 8.38%. We also show visible improvements aroundobject boundaries as well as a greater ability to generalize across datasets.Code is available athttps://github.com/duke-vision/semantic-unsup-flow-release.

SemARFlow: Injecting Semantics into Unsupervised Optical Flow Estimation for Autonomous Driving