Demystifying Catastrophic Forgetting in Two-Stage Incremental Object Detector

TL;DR

This paper investigates catastrophic forgetting in incremental object detection, revealing that it mainly affects the RoI Head classifier, and proposes a novel framework NSGP-RePRE that combines prototype replay and gradient projection to mitigate this issue, achieving state-of-the-art results.

Contribution

The paper introduces a component-specific analysis of catastrophic forgetting in IOD and proposes the NSGP-RePRE framework combining prototype replay and null space gradient projection.

Findings

RoI Head classifier is most affected by forgetting.

NSGP-RePRE achieves state-of-the-art results on Pascal VOC and MS COCO.

Component-specific analysis informs better mitigation strategies.

Abstract

Catastrophic forgetting is a critical chanllenge for incremental object detection (IOD). Most existing methods treat the detector monolithically, relying on instance replay or knowledge distillation without analyzing component-specific forgetting. Through dissection of Faster R-CNN, we reveal a key insight: Catastrophic forgetting is predominantly localized to the RoI Head classifier, while regressors retain robustness across incremental stages. This finding challenges conventional assumptions, motivating us to develop a framework termed NSGP-RePRE. Regional Prototype Replay (RePRE) mitigates classifier forgetting via replay of two types of prototypes: coarse prototypes represent class-wise semantic centers of RoI features, while fine-grained prototypes model intra-class variations. Null Space Gradient Projection (NSGP) is further introduced to eliminate prototype-feature misalignment by updating the feature extractor in directions orthogonal to subspace of old inputs via gradient projection, aligning RePRE with incremental learning dynamics. Our simple yet effective design allows NSGP-RePRE to achieve state-of-the-art performance on the Pascal VOC and MS COCO datasets under various settings. Our work not only advances IOD methodology but also provide pivotal insights for catastrophic forgetting mitigation in IOD. Code is available at \href{https://github.com/fanrena/NSGP-RePRE}{https://github.com/fanrena/NSGP-RePRE} .