Adaptive Class Suppression Loss for Long-Tail Object Detection
This addresses the challenge of detecting rare objects in large vocabulary datasets, which is crucial for real-world applications like autonomous driving and surveillance, but is incremental as it builds on existing long-tail methods.
The paper tackles the problem of long-tail distribution in object detection by proposing an Adaptive Class Suppression Loss (ACSL) to improve training consistency and discrimination for rare categories, achieving improvements of 5.18% and 5.2% on LVIS and Open Images datasets.
To address the problem of long-tail distribution for the large vocabulary object detection task, existing methods usually divide the whole categories into several groups and treat each group with different strategies. These methods bring the following two problems. One is the training inconsistency between adjacent categories of similar sizes, and the other is that the learned model is lack of discrimination for tail categories which are semantically similar to some of the head categories. In this paper, we devise a novel Adaptive Class Suppression Loss (ACSL) to effectively tackle the above problems and improve the detection performance of tail categories. Specifically, we introduce a statistic-free perspective to analyze the long-tail distribution, breaking the limitation of manual grouping. According to this perspective, our ACSL adjusts the suppression gradients for each sample of each class adaptively, ensuring the training consistency and boosting the discrimination for rare categories. Extensive experiments on long-tail datasets LVIS and Open Images show that the our ACSL achieves 5.18% and 5.2% improvements with ResNet50-FPN, and sets a new state of the art. Code and models are available at https://github.com/CASIA-IVA-Lab/ACSL.