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

Learning from imperfect data becomes an issue in many industrial applicationsafter the research community has made profound progress in supervised learningfrom perfectly annotated datasets. The purpose of the Learning from ImperfectData (LID) workshop is to inspire and facilitate the research in developingnovel approaches that would harness the imperfect data and improve thedata-efficiency during training. A massive amount of user-generated datanowadays available on multiple internet services. How to leverage those andimprove the machine learning models is a high impact problem. We organize thechallenges in conjunction with the workshop. The goal of these challenges is tofind the state-of-the-art approaches in the weakly supervised learning settingfor object detection, semantic segmentation, and scene parsing. There are threetracks in the challenge, i.e., weakly supervised semantic segmentation (Track1), weakly supervised scene parsing (Track 2), and weakly supervised objectlocalization (Track 3). In Track 1, based on ILSVRC DET, we provide pixel-levelannotations of 15K images from 200 categories for evaluation. In Track 2, weprovide point-based annotations for the training set of ADE20K. In Track 3,based on ILSVRC CLS-LOC, we provide pixel-level annotations of 44,271 imagesfor evaluation. Besides, we further introduce a new evaluation metric proposedby \cite{zhang2020rethinking}, i.e., IoU curve, to measure the quality of thegenerated object localization maps. This technical report summarizes thehighlights from the challenge. The challenge submission server and theleaderboard will continue to open for the researchers who are interested in it.More details regarding the challenge and the benchmarks are available athttps://lidchallenge.github.io

LID 2020: The Learning from Imperfect Data Challenge Results