Decoupling Classifier for Boosting Few-shot Object Detection and Instance Segmentation

TL;DR

This paper introduces a simple decoupling method for the classifier in few-shot object detection and segmentation, effectively reducing bias and improving performance without extra computational cost.

Contribution

It proposes a decoupled classifier with two heads to better handle positive and noisy negative samples caused by missing labels in FSOD and FSIS.

Findings

Outperforms baseline and state-of-the-art methods on PASCAL VOC and MS-COCO.

Effectively mitigates bias caused by missing labels in few-shot scenarios.

No additional computation or parameters required.

Abstract

This paper focus on few-shot object detection~(FSOD) and instance segmentation~(FSIS), which requires a model to quickly adapt to novel classes with a few labeled instances. The existing methods severely suffer from bias classification because of the missing label issue which naturally exists in an instance-level few-shot scenario and is first formally proposed by us. Our analysis suggests that the standard classification head of most FSOD or FSIS models needs to be decoupled to mitigate the bias classification. Therefore, we propose an embarrassingly simple but effective method that decouples the standard classifier into two heads. Then, these two individual heads are capable of independently addressing clear positive samples and noisy negative samples which are caused by the missing label. In this way, the model can effectively learn novel classes while mitigating the effects of noisy negative samples. Without bells and whistles, our model without any additional computation cost and parameters consistently outperforms its baseline and state-of-the-art by a large margin on PASCAL VOC and MS-COCO benchmarks for FSOD and FSIS tasks. The Code is available at https://csgaobb.github.io/Projects/DCFS.