Replay Consolidation with Label Propagation for Continual Object Detection

TL;DR

This paper introduces RCLPOD, a novel replay consolidation method with label propagation for continual object detection, addressing challenges like missing annotations and class imbalance, and demonstrating superior performance on benchmarks.

Contribution

It proposes a new replay-based approach that enhances sample quality and class balance using label propagation, differing from traditional distillation methods in continual object detection.

Findings

RCLPOD outperforms existing methods on VOC and COCO benchmarks.

The approach is compatible with modern architectures like YOLOv8.

It effectively handles class imbalance and missing annotations in continual learning.

Abstract

Continual Learning (CL) aims to learn new data while remembering previously acquired knowledge. In contrast to CL for image classification, CL for Object Detection faces additional challenges such as the missing annotations problem. In this scenario, images from previous tasks may contain instances of unknown classes that could reappear as labeled in future tasks, leading to task interference in replay-based approaches. Consequently, most approaches in the literature have focused on distillation-based techniques, which are effective when there is a significant class overlap between tasks. In our work, we propose an alternative to distillation-based approaches with a novel approach called Replay Consolidation with Label Propagation for Object Detection (RCLPOD). RCLPOD enhances the replay memory by improving the quality of the stored samples through a technique that promotes class balance while also improving the quality of the ground truth associated with these samples through a technique called label propagation. RCLPOD outperforms existing techniques on well-established benchmarks such as VOC and COC. Moreover, our approach is developed to work with modern architectures like YOLOv8, making it suitable for dynamic, real-world applications such as autonomous driving and robotics, where continuous learning and resource efficiency are essential.