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

Object detection is an important task in computer vision which serves a lotof real-world applications such as autonomous driving, surveillance androbotics. Along with the rapid thrive of large-scale data, numerousstate-of-the-art generalized object detectors (e.g. Faster R-CNN, YOLO, SSD)were developed in the past decade. Despite continual efforts in modelmodification and improvement in training strategies to boost detectionaccuracy, there are still limitations in performance of detectors when it comesto specialized datasets with uneven object class distributions. This originatesfrom the common usage of Cross Entropy loss function for object classificationsub-task that simply ignores the frequency of appearance of object class duringtraining, and thus results in lower accuracies for object classes with fewernumber of samples. Class-imbalance in general machine learning has been widelystudied, however, little attention has been paid on the subject of objectdetection. In this paper, we propose to explore and overcome such problem byapplication of several weighted variants of Cross Entropy loss, for examplesBalanced Cross Entropy, Focal Loss and Class-Balanced Loss Based on EffectiveNumber of Samples to our object detector. Experiments with BDD100K (a highlyclass-imbalanced driving database acquired from on-vehicle cameras capturingmostly Car-class objects and other minority object classes such as Bus, Personand Motor) have proven better class-wise performances of detector trained withthe afore-mentioned loss functions.

Resolving Class Imbalance in Object Detection with Weighted Cross Entropy Losses