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

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 a 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://github.com/gaobb/DCFS.

PDF

Datasets


Task Dataset Model Metric Name Metric Value Global Rank Benchmark
Few-Shot Object Detection MS-COCO (10-shot) DCFS AP 19.5 # 9
Few-Shot Object Detection MS-COCO (1-shot) DCFS AP 10.0 # 4
AP 8.1 # 6
Few-Shot Object Detection MS-COCO (30-shot) DCFS AP 22.7 # 10

Methods


No methods listed for this paper. Add relevant methods here