Towards Generalized and Incremental Few-Shot Object Detection

TL;DR

This paper introduces a novel incremental few-shot object detection method that effectively addresses catastrophic forgetting and overfitting, enabling continual learning of new classes with limited samples.

Contribution

The paper proposes a Double-Branch Framework and progressive model updating to improve incremental few-shot detection, a novel approach for continual learning with few samples.

Findings

Significantly improves detection accuracy on Pascal VOC and MS-COCO.

Effectively retains old knowledge while learning new classes.

Addresses catastrophic forgetting and overfitting in incremental few-shot detection.

Abstract

Real-world object detection is highly desired to be equipped with the learning expandability that can enlarge its detection classes incrementally. Moreover, such learning from only few annotated training samples further adds the flexibility for the object detector, which is highly expected in many applications such as autonomous driving, robotics, etc. However, such sequential learning scenario with few-shot training samples generally causes catastrophic forgetting and dramatic overfitting. In this paper, to address the above incremental few-shot learning issues, a novel Incremental Few-Shot Object Detection (iFSOD) method is proposed to enable the effective continual learning from few-shot samples. Specifically, a Double-Branch Framework (DBF) is proposed to decouple the feature representation of base and novel (few-shot) class, which facilitates both the old-knowledge retention and new-class adaption simultaneously. Furthermore, a progressive model updating rule is carried out to preserve the long-term memory on old classes effectively when adapt to sequential new classes. Moreover, an inter-task class separation loss is proposed to extend the decision region of new-coming classes for better feature discrimination. We conduct experiments on both Pascal VOC and MS-COCO, which demonstrate that our method can effectively solve the problem of incremental few-shot detection and significantly improve the detection accuracy on both base and novel classes.