Jia Lin

Jia Lin

Auto-generated profile, based on publicly available information

Jia Lin

School of Automation, Hangzhou Dianzi University, Hangzhou, China

Research Focus & Metrics

Academic Career Length

2Years (2025-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
11 March 2025 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
JHJiacheng Huang
Jiacheng Huang
YWYong Wu
Yong Wu
XZXiaofei Zhou
Xiaofei Zhou
JLJia Lin
Jia Lin
ZCZhangping Chen
Zhangping Chen
JZJiyong Zhang
Jiyong Zhang
In recent years, few-shot semantic segmentation of strip steel surface defects (S3D) has posed significant challenges, primarily due to the wide diversity of scenarios and the inherent complexity of defects. However, existing semantic segmentation methods have not effectively addressed these issues. In this paper, we propose a few-shot segmentation method called the Multiscale Adaptive Prototype Transformer Network (MAPTNet), which aims to integrate multi-scale feature aggregation and enhance the adaptability of defect detection across diverse and complex defect scenarios. Specifically, we introduce an Adaptive Prototype Transformer (APT) module to locate defects better and generate more refined feature representations. This module effectively explores the relationships between support and query sets and transforms them into optimal prototypes. Additionally, through hierarchical feature fusion to capture more detailed and contextually relevant S3D information, our MAPTNet improves the discriminability of defects at multiple feature scales. To further strengthen the learning process across different feature layers, our MAPTNet incorporates a deep dual supervision mechanism, facilitating effective optimization at both intermediate and final stages. Extensive experiments on the FSSD-12, Surface Defects-4i, and ESDIs-SOD datasets demonstrate that our network achieves state-of-the-art performance, significantly outperforming existing approaches. The code is available at https://github.com/hhhjjc/MAPTNet.

19 February 2026 in IEEE Transactions on Image Processing
Verified
Journal Article
Closed access
XZXiaofei Zhou
Xiaofei Zhou
JLJia Lin
Jia Lin
DCDongmei Chen
Dongmei Chen
DLDeyang Liu
Deyang Liu
JZJiyong Zhang
Jiyong Zhang
RCRunmin Cong
Runmin Cong
Multi-modal few-shot semantic segmentation (FSS) aims to perform dense prediction from multiple modality images including visible image, depth image, and thermal image with a few annotated samples. However, some efforts treat the three modality information equally, where they don’t incorporate the inherent differences among multiple modalities. Besides, the objects vary in size greatly, and the cutting-edge matching paradigms fail to establish an effective support-query connection. Therefore, we propose a novel scale-invariant feature matching network (i.e., SFM-Net), which consists of an encoder, a feature matching block, a feature elevation block, and a decoder, to conduct visible-depth-thermal (V-D-T) few-shot semantic segmentation. Firstly, in the encoder part, after the extraction of multi-level initial features, we fuse each level’s RGB feature and thermal feature, yielding the support features and the query features. Secondly, in the feature matching block, a pixel-to-patch cross-attention (PTPCA) module is deployed to explore the correlation between each level’s support feature and the query feature, where the pixel-to-patch pooling (PTP-pool) units are designed to build scale-invariant relationships, generating the coarse mask for the query image. Thirdly, in the feature elevation block, we employ the prior-related fusion (PF) module to integrate the depth image with a coarse mask via the cross-attention mechanism, yielding the enhanced coarse prediction result, which is further aggregated in a bottom-up way. Finally, in the decoder, we deploy a reverse attention (RA) unit to gradually explore the complementarity between object internal regions and spatial details, and further generate the final segmentation results via conventional convolution layers. Extensive experiments are conducted on the VDT-2048-5i dataset, and the results show that our model outperforms the state-of-the-art methods with a large margin.