ETRI-Knowledge Sharing Plaform

ENGLISH

성과물

논문 검색
구분 SCI
연도 ~ 키워드

상세정보

학술지 Dynamic Residual Filtering with Laplacian Pyramid for Instance Segmentation
Cited 0 time in scopus Download 7 time Share share facebook twitter linkedin kakaostory
저자
송민수, 엄기문, 이희경, 서정일, 김원준
발행일
202311
출처
IEEE Transactions on Multimedia, v.25, pp.1-12
ISSN
1520-9210
출판사
IEEE
DOI
https://dx.doi.org/10.1109/TMM.2022.3215306
협약과제
21HH4900, [전문연구실] 이머시브 미디어 전문연구실, 서정일
초록
Various studies have been conducted on instance segmentation and made great strides over the past few years. Most recently, instance-specific mask generation via dynamic kernel predictions has shown the significant performance improvement even without bounding boxes as well as anchors. However, this scheme still does not fully consider dynamic properties since the size of the receptive field is not enough to cover the spatially-meaningful range due to memory limitations. Furthermore, the single-fused feature often fails to grasp complicated boundaries for objects of different sizes. In this paper, we propose the dynamic residual filtering method with the Laplacian pyramid, which separately restores the global layout and local boundaries of instance masks. Specifically, we firstly apply the Laplacian pyramid-based decomposition scheme to features encoded from the backbone and subsequently restore sub-band mask residuals from coarse to fine pyramid levels. To do this, we design spatially-aware convolution filters to progressively capture the residual form of mask features at each level of the Laplacian pyramid while holding deformable receptive fields with dynamic offset information. This is fairly desirable since global and local properties of mask features can be accurately restored with keeping the spatial flexibility through the invertible process of the Laplacian reconstruction. Experimental results on the COCO dataset demonstrate that our proposed method achieves the state-of-the-art performance, i.e., 42.7% AP. The code and model are publicly available at: https://github.com/tjqansthd/LapMask.
KSP 제안 키워드
Art performance, Bounding Box, Convolution filters, Decomposition scheme, Different sizes, Dynamic offset, Dynamic properties, Filtering method, Global and local, Laplacian pyramid, Local properties