CAT: A Simple yet Effective Cross-Attention Transformer for One-Shot Object Detection

Simple item page
dc.contributor.authorLin, W.D.
dc.contributor.authorDeng, Y.Y.
dc.contributor.authorGao, Y.
dc.contributor.authorWang, N.
dc.contributor.authorLiu, L.Q.
dc.contributor.authorZhang, L.
dc.contributor.authorWang, P.
dc.date.issued2024
dc.description.abstractGiven a query patch from a novel class, one-shot object detection aims to detect all instances of this class in a target image through the semantic similarity comparison. However, due to the extremely limited guidance in the novel class as well as the unseen appearance difference between the query and target instances, it is difficult to appropriately exploit their semantic similarity and generalize well. To mitigate this problem, we present a universal Cross-Attention Transformer (CAT) module for accurate and efficient semantic similarity comparison in one-shot object detection. The proposed CAT utilizes the transformer mechanism to comprehensively capture bi-directional correspondence between any paired pixels from the query and the target image, which empowers us to sufficiently exploit their semantic characteristics for accurate similarity comparison. In addition, the proposed CAT enables feature dimensionality compression for inference speedup without performance loss. Extensive experiments on three object detection datasets MS-COCO, PASCAL VOC and FSOD under the one-shot setting demonstrate the effectiveness and efficiency of our model, e.g., it surpasses CoAE, a major baseline in this task, by 1.0% in average precision (AP) on MS-COCO and runs nearly 2.5 times faster.
dc.description.statementofresponsibilityWei-Dong Lin (林蔚东), Yu-Yan Deng (邓玉岩), Yang Gao (高　扬), Ning Wang (王　宁), Ling-Qiao Liu (刘凌峤), Lei Zhang (张　磊), and Peng Wang (王　鹏)
dc.identifier.citationJournal of Computer Science and Technology, 2024; 39(2):460-471
dc.identifier.doi10.1007/s11390-024-1743-6
dc.identifier.issn1000-9000
dc.identifier.issn1860-4749
dc.identifier.urihttps://hdl.handle.net/2440/147946
dc.language.isoen
dc.publisherSpringer
dc.relation.granthttp://purl.org/au-research/grants/arc/2021JCW-03
dc.rights© Institute of Computing Technology, Chinese Academy of Sciences 2024
dc.source.urihttp://dx.doi.org/10.1007/s11390-024-1743-6
dc.subjectone-shot object detection; Transformer; attention mechanism
dc.titleCAT: A Simple yet Effective Cross-Attention Transformer for One-Shot Object Detection
dc.typeJournal article
pubs.publication-statusPublished

Files

Collections

Research Outputs