Please use this identifier to cite or link to this item:
Title: Efficient few-shot object detection via knowledge inheritance
Authors: Yang, Ze
Zhang, Chi
Li, Ruibo
Xu, Yi
Lin, Guosheng
Keywords: Engineering::Computer science and engineering
Issue Date: 2023
Source: Yang, Z., Zhang, C., Li, R., Xu, Y. & Lin, G. (2023). Efficient few-shot object detection via knowledge inheritance. IEEE Transactions On Image Processing, 32, 321-334.
Project: AISG-RP-2018-003 
Journal: IEEE Transactions on Image Processing 
Abstract: Few-shot object detection (FSOD), which aims at learning a generic detector that can adapt to unseen tasks with scarce training samples, has witnessed consistent improvement recently. However, most existing methods ignore the efficiency issues, e.g., high computational complexity and slow adaptation speed. Notably, efficiency has become an increasingly important evaluation metric for few-shot techniques due to an emerging trend toward embedded AI. To this end, we present an efficient pretrain-transfer framework (PTF) baseline with no computational increment, which achieves comparable results with previous state-of-the-art (SOTA) methods. Upon this baseline, we devise an initializer named knowledge inheritance (KI) to reliably initialize the novel weights for the box classifier, which effectively facilitates the knowledge transfer process and boosts the adaptation speed. Within the KI initializer, we propose an adaptive length re-scaling (ALR) strategy to alleviate the vector length inconsistency between the predicted novel weights and the pretrained base weights. Finally, our approach not only achieves the SOTA results across three public benchmarks, i.e., PASCAL VOC, COCO and LVIS, but also exhibits high efficiency with 1.8-100× faster adaptation speed against the other methods on COCO/LVIS benchmark during few-shot transfer. To our best knowledge, this is the first work to consider the efficiency problem in FSOD. We hope to motivate a trend toward powerful yet efficient few-shot technique development. The codes are publicly available at
ISSN: 1057-7149
DOI: 10.1109/TIP.2022.3228162
Schools: School of Computer Science and Engineering 
Rights: © 2022 IEEE. All rights reserved.
Fulltext Permission: none
Fulltext Availability: No Fulltext
Appears in Collections:SCSE Journal Articles

Citations 20

Updated on May 24, 2024

Web of ScienceTM
Citations 50

Updated on Oct 30, 2023

Page view(s)

Updated on May 24, 2024

Google ScholarTM




Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.