A MBGD enhancement method for imbalance smoothing

Xusheng Ai, Victor S. Sheng, Chunhua Li

Research output: Contribution to journalArticlepeer-review


This paper addresses foreground-foreground imbalance in object detection. Firstly, we introduce Mini-batch Stochastic Gradient Descent (MBGD) with YOLO and the foreground-foreground imbalance problem. Then T-distribution is devised and proved to smoothen the imbalanced distribution and allocate at least a representative for each class. Furthermore, Mini-Batch Imbalance Smoothing method (MB-IS) is proposed to address the foreground-foreground imbalance by following T-distribution and proportionally assigning class weights in a mini-batch. Finally, Extensive experiments on our own transaction dataset and VOC2007 dataset demonstrate the superiority of MB-IS with certain mini-batch size.

Original languageEnglish
JournalMultimedia Tools and Applications
StateAccepted/In press - 2022


  • Foreground-foreground imbalance
  • Mini-batch stochastic gradient descent
  • Object detection
  • YOLO


Dive into the research topics of 'A MBGD enhancement method for imbalance smoothing'. Together they form a unique fingerprint.

Cite this