Equalization Loss for Long-Tailed Object Recognition

Abstract

Object recognition techniques using convolutional neural networks (CNN) have achieved great success. However, state-of-the-art object detection methods still perform poorly on large vocabulary and long-tailed datasets, e.g. LVIS. In this work, we analyze this problem from a novel perspective: each positive sample of one category can be seen as a negative sample for other categories, making the tail categories receive more discouraging gradients. Based on it, we propose a simple but effective loss, named equalization loss, to tackle the problem of long-tailed rare categories by simply ignoring those gradients for rare categories. The equalization loss protects the learning of rare categories from being at a disadvantage during the network parameter updating. Thus the model is capable of learning better discriminative features for objects of rare classes. Without any bells and whistles, our method achieves AP gains of 4.1% and 4.8% for the rare and common categories on the challenging LVIS benchmark, compared to the Mask R-CNN baseline. With the utilization of the effective equalization loss, we finally won the 1st place in the LVIS Challenge 2019. Code has been made available at: https: //github.com/tztztztztz/eql.detectron2

CVPR 2020. Winner of LVIS Challenge 2019. Code has been available at https: //github.com/tztztztztz/eql.detectron2

References in corpus (10)

• SMOTE: Synthetic Minority Over-sampling Technique
• Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
• Distributed Representations of Words and Phrases and their Compositionality
• Improved Regularization of Convolutional Neural Networks with Cutout
• Accurate, Large Minibatch SGD: Training ImageNet in 1 Hour
• Squeeze-and-Excitation Networks
• Deeply-Supervised Nets
• The Open Images Dataset V4: Unified image classification, object detection, and visual relationship detection at scale
• Meta-Weight-Net: Learning an Explicit Mapping For Sample Weighting
• Decoupling Representation and Classifier for Long-Tailed Recognition

Cited by in corpus (8)

• Long-Tailed Classification by Keeping the Good and Removing the Bad Momentum Causal Effect
• CReST: A Class-Rebalancing Self-Training Framework for Imbalanced Semi-Supervised Learning
• TransAction: ICL-SJTU Submission to EPIC-Kitchens Action Anticipation Challenge 2021
• Image-Level or Object-Level? A Tale of Two Resampling Strategies for Long-Tailed Detection
• Disentangling Sampling and Labeling Bias for Learning in Large-Output Spaces
• Graph Classification by Mixture of Diverse Experts
• Class-Wise Difficulty-Balanced Loss for Solving Class-Imbalance
• Simple multi-dataset detection