Few-Shot Batch Incremental Road Object Detection via Detector Fusion

Anuj Tambwekar, Kshitij Agrawal, Anay Majee, Anbumani Subramanian

Abstract: Incremental few-shot learning has emerged as a new and challenging area in deep learning, whose objective is to train deep learning models using very few samples of new class data, and none of the old class data. In this work we tackle the problem of batch incremental few-shot road object detection using data from the India Driving Dataset (IDD). Our approach, DualFusion, combines object detectors in a manner that allows us to learn to detect rare objects with very limited data, all without severely degrading the performance of the de4tector on the abundant classes. In the IDD OpenSet incremental few-shot detection task, we achieve a mAP <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">50</inf> score of 40.0 on the base classes and an overall mAP <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">50</inf> score of 38.8, both of which are the highest to date. In the COCO batch incremental few-shot detection task, we achieve a novel AP score of 9.9, surpassing the state-of-the-art novel class performance on the same by over 6.6 times.