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

We study on weakly-supervised object detection (WSOD) which plays a vitalrole in relieving human involvement from object-level annotations. Predominantworks integrate region proposal mechanisms with convolutional neural networks(CNN). Although CNN is proficient in extracting discriminative local features,grand challenges still exist to measure the likelihood of a bounding boxcontaining a complete object (i.e., "objectness"). In this paper, we propose anovel WSOD framework with Objectness Distillation (i.e., WSOD^2) by designing atailored training mechanism for weakly-supervised object detection. Multipleregression targets are specifically determined by jointly considering bottom-up(BU) and top-down (TD) objectness from low-level measurement and CNNconfidences with an adaptive linear combination. As bounding box regression canfacilitate a region proposal learning to approach its regression target withhigh objectness during training, deep objectness representation learned frombottom-up evidences can be gradually distilled into CNN by optimization. Weexplore different adaptive training curves for BU/TD objectness, and show thatthe proposed WSOD^2 can achieve state-of-the-art results.

WSOD^2: Learning Bottom-up and Top-down Objectness Distillation for Weakly-supervised Object Detection