基于记忆库和后处理方法解决长尾实例分割问题

Method for long-tailed instance segmentation based on memory bank and confidence calibration

数据的长尾特性一直是解决计算机视觉问题的一大挑战,为了应对这一挑战在实例分割任务中带来的困难,通常引入图像重采样作为一种简单有效的方法。然而一幅图像中可能存在多个类,仅通过图像重采样不足以在类别的数据量层面上获得足够均衡的分布,因此提出一种以目标为中心的记忆库后处理方法。首先引入记忆库模型,设立以目标为单位的存储策略,解决各类别数量不均衡问题;其次使用一种后处理校准方法,通过调整每个类的置信度得分提升稀有类和普通类的预测值。在公开数据集LVIS上的实验验证了所提方法的有效性,并且相比EQL方法,准确率能够提升2.2%。

Long-tail characteristic of data has always been a great challenge in solving computer vision problems.To solve the difficulties brought to instance segmentation,image re-sampling,a simple and efficient method has always been used.However,there may be many classes in one image,it’s hard to balance different classes on the data level.This paper proposed a object-centric post-processable memory bank method.Firstly,it introduced a memory bank model and set up an object-based storage policy to address the number imbalance of categories.Then,in order to increase the predicted score for tail classes and common classes,it used a post-processing calibration to adjust the confidence score of each class.This paper verified the effectiveness of the proposed method by experiments on LVIS dataset.The accuracy can be improved by 2.2%compared to EQL method.