Online Learning via Memory: Retrieval-Augmented Detector Adaptation
This addresses the challenge of domain adaptation for object detection in computer vision, offering a training-free solution that is incremental in its approach.
The paper tackles the problem of adapting object detection models to new domains without retraining by introducing a retrieval-augmented classification module and memory bank, achieving significant performance improvements over baselines with minimal memory usage (e.g., 10 images per category) and no training.
This paper presents a novel way of online adapting any off-the-shelf object detection model to a novel domain without retraining the detector model. Inspired by how humans quickly learn knowledge of a new subject (e.g., memorization), we allow the detector to look up similar object concepts from memory during test time. This is achieved through a retrieval augmented classification (RAC) module together with a memory bank that can be flexibly updated with new domain knowledge. We experimented with various off-the-shelf open-set detector and close-set detectors. With only a tiny memory bank (e.g., 10 images per category) and being training-free, our online learning method could significantly outperform baselines in adapting a detector to novel domains.