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/

Zendy Plus

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

Zendy Tools

Zendy Open

Image restoration algorithms such as super resolution (SR) are indispensablepre-processing modules for object detection in low quality images. Most ofthese algorithms assume the degradation is fixed and known a priori. However,in practical, either the real degradation or optimal up-sampling ratio rate isunknown or differs from assumption, leading to a deteriorating performance forboth the pre-processing module and the consequent high-level task such asobject detection. Here, we propose a novel self-supervised framework to detectobjects in degraded low resolution images. We utilizes the downsamplingdegradation as a kind of transformation for self-supervised signals to explorethe equivariant representation against various resolutions and otherdegradation conditions. The Auto Encoding Resolution in Self-supervision(AERIS) framework could further take the advantage of advanced SR architectureswith an arbitrary resolution restoring decoder to reconstruct the originalcorrespondence from the degraded input image. Both the representation learningand object detection are optimized jointly in an end-to-end training fashion.The generic AERIS framework could be implemented on various mainstream objectdetection architectures with different backbones. The extensive experimentsshow that our methods has achieved superior performance compared with existingmethods when facing variant degradation situations. Code would be released athttps://github.com/cuiziteng/ECCV_AERIS.

Exploring Resolution and Degradation Clues as Self-supervised Signal for Low Quality Object Detection