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

The majority of current object detectors lack context: class predictions aremade independently from other detections. We propose to incorporate context inobject detection by post-processing the output of an arbitrary detector torescore the confidences of its detections. Rescoring is done by conditioning oncontextual information from the entire set of detections: their confidences,predicted classes, and positions. We show that AP can be improved by simplyreassigning the detection confidence values such that true positives thatsurvive longer (i.e., those with the correct class and large IoU) are scoredhigher than false positives or detections with small IoU. In this setting, weuse a bidirectional RNN with attention for contextual rescoring and introduce atraining target that uses the IoU with ground truth to maximize AP for thegiven set of detections. The fact that our approach does not require access tovisual features makes it computationally inexpensive and agnostic to thedetection architecture. In spite of this simplicity, our model consistentlyimproves AP over strong pre-trained baselines (Cascade R-CNN and Faster R-CNNwith several backbones), particularly by reducing the confidence of duplicatedetections (a learned form of non-maximum suppression) and removingout-of-context objects by conditioning on the confidences, classes, positions,and sizes of the co-occurrent detections. Code is available athttps://github.com/LourencoVazPato/seeing-without-looking/

Seeing without Looking: Contextual Rescoring of Object Detections for AP Maximization