A5033395835- Advanced Image and Video Retrieval Techniques
- Video Surveillance and Tracking Methods
- Advanced Neural Network Applications
- Domain Adaptation and Few-Shot Learning
- Human Pose and Action Recognition
- Multimodal Machine Learning Applications
- Image Retrieval and Classification Techniques
- Anomaly Detection Techniques and Applications
- Remote-Sensing Image Classification
- Gait Recognition and Analysis
- Salivary Gland Tumors Diagnosis and Treatment
- Head and Neck Surgical Oncology
- Visual Attention and Saliency Detection
- Oral and Maxillofacial Pathology
- Neural Networks and Applications
- RNA modifications and cancer
- Face recognition and analysis
- Cancer-related gene regulation
- Network Security and Intrusion Detection
- Bone Tumor Diagnosis and Treatments
- Osteoarthritis Treatment and Mechanisms
- Autophagy in Disease and Therapy
- Image and Signal Denoising Methods
- Cancer-related molecular mechanisms research
- Facial Trauma and Fracture Management
Northwestern Polytechnical University
2017-2025
Shandong First Medical University
2021-2025
Shandong Provincial Hospital
2014-2025
Northwestern Polytechnic University
2023
Ningbo Polytechnic
2023
Shandong University
2011-2021
Xi'an Jiaotong University
2014-2017
Government Medical College
2015
University of Calgary
2015
MIT Lincoln Laboratory
2015
Nowadays, with the rapid development of consumer Unmanned Aerial Vehicles (UAVs), visual surveillance by utilizing UAV platform has been very attractive. Most research works for captured data are mainly focused on tasks object detection and tracking. However, limited attention paid to task person Re-identification (ReID) which widely studied in ordinary cameras fixed emplacements. In this paper, facilitate ReID aerial imagery, we collect a large scale airborne dataset named as Person Imagery...
Head and neck squamous cell carcinoma (HNSCC) is the six leading cancer by incidence worldwide. The 5-year survival rate of HNSCC patients remains less than 65% due to lack symptoms in early stage. Hence, biomarkers which can improve detection should clinical outcome. Gene expression profiles (GSE6631, GSE58911) Cancer Genome Atlas (TCGA) data were used for integrated bioinformatics analysis; differentially expressed genes (DEGs) then subjected functional pathway enrichment analysis,...
Abstract Background Due to the lack of research on pathological mechanism temporomandibular joint osteoarthritis (TMJOA), there are few effective treatment measures in clinic. In recent years, microRNAs (miRs) have been demonstrated play an important role pathogenesis (OA) by regulating a variety target genes, and latest evidence shows that miR-21-5p is specifically overexpressed OA. The purpose this project was clarify whether can regulate TMJOA process targeting Spry1. Methods induced...
In this work, we construct a large-scale dataset for vehicle re-identification (ReID), which contains 137k images of 13k instances captured by UAV-mounted cameras. To our knowledge, it is the largest UAV-based ReID dataset. increase intra-class variation, each at least two UAVs different locations, with diverse view-angles and flight-altitudes. We manually label variety attributes, including type, color, skylight, bumper, spare tire luggage rack. Furthermore, image, annotator also required...
The problem of cross-modality person re-identification has been receiving increasing attention recently, due to its practical significance. Motivated by the fact that human usually attend difference when they compare two similar objects, we propose a dual-path feature learning framework which preserves intrinsic spatial structures and attends input image pairs. Our is composed main components: Dual-path Spatial-structure-preserving Common Space Network (DSCSN) Contrastive Correlation (CCN)....
For weakly supervised anomaly detection, most existing work is limited to the problem of inadequate video representation due inability modeling long-term contextual information. To solve this, we propose a novel adaptive graph convolutional network (WAGCN) model complex relationship among segments. By which, fully consider influence other segments on current one when generating probability score for each segment. Firstly, combine temporal consistency as well feature similarity construct...
With the emergence of large pre-trained vison-language model like CLIP, transferable representations can be adapted to a wide range downstream tasks via prompt tuning. Prompt tuning tries probe beneficial information for from general knowledge stored in model. A recently proposed method named Context Optimization (CoOp) introduces set learnable vectors as text language side. However, alone only adjust synthesized "classifier", while computed visual features image encoder not affected , thus...
Oral microbiota dysbiosis is associated with the occurrence and progression of oral cancer. To investigate association between risk squamous cell carcinoma (OSCC), we identified microbial composition paired tumor (TT)/normal paracancerous tissues (NPT) saliva (TS) samples in OSCC patients through 16S rRNA gene sequencing. A total 22 phyla, 321 genera, 869 species were samples. Paired comparisons revealed significant differences TT, NPT, TS groups, genus Filifactor significantly enriched TT....
The Rho-associated protein kinase (ROCK) inhibitor Y-27632 is a potential immunotherapeutic agent for cancer treatment. blocks the growth and migration of oral squamous cell carcinoma (OSCC) CAL-27 cells. However, detailed studies on underlying mechanisms have not yet been reported. We investigated effects proliferation, migration, invasion OSCC cells (CAL-27, SCC-4, SCC-9) using Cell Counting Kit-8 assay, ethynyl-2'-deoxyuridine staining, scratch, transwell assay in vitro. Next, ROCK1/2 was...
A critical challenge for multi-modal Object Re-Identification (ReID) is the effective aggregation of complementary information to mitigate illumination issues. State-of-the-art methods typically employ complex and highly-coupled architectures, which unavoidably result in heavy computational costs. Moreover, significant distribution gap among different image spectra hinders joint representation multimodal features. In this paper, we propose a framework named as PromptMA establish...
Fusion of a panchromatic (PAN) image and corresponding multispectral (MS) is also known as pansharpening, which aims to combine abundant spatial details PAN spectral information MS images. Due the absence high-resolution images, available deep-learning-based methods usually follow paradigm training at reduced resolution testing both full resolution. When taking original images inputs, they always obtain sub-optimal results due scale variation. In this paper, we propose explore...
Abstract Objectives/Hypothesis: Is the severity of acute oral mucositis in patients who receive postoperative intensity‐modulated radiotherapy (PO‐IMRT) for tongue squamous cell carcinoma (SCC) reduced by sparing mucosa outside planning target volume (PTV)? Study Design: Prospective, randomized trial. Methods: Forty‐eight with SCC received PO‐IMRT at our institution were to two groups: oral‐sparing (OR‐SP) group and oral‐unsparing (OR‐USP) group. For OR‐SP (n = 24), PTV was spared....
Person re-identification (ReID), which aims at matching individuals across non-overlapping cameras, has attracted much attention in the field of computer vision due to its research significance and potential applications. Triplet loss-based CNN models have been very successful for person ReID, optimize feature embedding space such that distances between samples with same identity are shorter than those different identities. Researchers found hard triplets' mining is crucial success triplet...
Reversion‑inducing cysteine‑rich protein with Kazal motifs (RECK), a glycosylphosphatidylinositol‑anchored glycoprotein, inhibits the enzymatic activities of certain matrix metalloproteinases (MMPs). RECK has been studied in numerous human tumors, but expression salivary adenoid cystic carcinoma (SACC), and its correlation patient prognosis, never investigated thus far. In present study, MMP‑2 was evaluated two ACC cell lines 83 patients SACC. The results quantitative polymerase chain...