Teach YOLO to Remember: A Self-Distillation Approach for Continual Object Detection

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: De Monte, Riccardo, Pezze, Davide Dalle, Susto, Gian Antonio
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910861640073216
author De Monte, Riccardo
Pezze, Davide Dalle
Susto, Gian Antonio
author_facet De Monte, Riccardo
Pezze, Davide Dalle
Susto, Gian Antonio
contents Real-time object detectors like YOLO achieve exceptional performance when trained on large datasets for multiple epochs. However, in real-world scenarios where data arrives incrementally, neural networks suffer from catastrophic forgetting, leading to a loss of previously learned knowledge. To address this, prior research has explored strategies for Class Incremental Learning (CIL) in Continual Learning for Object Detection (CLOD), with most approaches focusing on two-stage object detectors. However, existing work suggests that Learning without Forgetting (LwF) may be ineffective for one-stage anchor-free detectors like YOLO due to noisy regression outputs, which risk transferring corrupted knowledge. In this work, we introduce YOLO LwF, a self-distillation approach tailored for YOLO-based continual object detection. We demonstrate that when coupled with a replay memory, YOLO LwF significantly mitigates forgetting. Compared to previous approaches, it achieves state-of-the-art performance, improving mAP by +2.1% and +2.9% on the VOC and COCO benchmarks, respectively.
format Preprint
id arxiv_https___arxiv_org_abs_2503_04688
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Teach YOLO to Remember: A Self-Distillation Approach for Continual Object Detection
De Monte, Riccardo
Pezze, Davide Dalle
Susto, Gian Antonio
Computer Vision and Pattern Recognition
Real-time object detectors like YOLO achieve exceptional performance when trained on large datasets for multiple epochs. However, in real-world scenarios where data arrives incrementally, neural networks suffer from catastrophic forgetting, leading to a loss of previously learned knowledge. To address this, prior research has explored strategies for Class Incremental Learning (CIL) in Continual Learning for Object Detection (CLOD), with most approaches focusing on two-stage object detectors. However, existing work suggests that Learning without Forgetting (LwF) may be ineffective for one-stage anchor-free detectors like YOLO due to noisy regression outputs, which risk transferring corrupted knowledge. In this work, we introduce YOLO LwF, a self-distillation approach tailored for YOLO-based continual object detection. We demonstrate that when coupled with a replay memory, YOLO LwF significantly mitigates forgetting. Compared to previous approaches, it achieves state-of-the-art performance, improving mAP by +2.1% and +2.9% on the VOC and COCO benchmarks, respectively.
title Teach YOLO to Remember: A Self-Distillation Approach for Continual Object Detection
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2503.04688