A5100435494- Advanced Vision and Imaging
- Advanced Image and Video Retrieval Techniques
- Advanced Neural Network Applications
- Image Enhancement Techniques
- Generative Adversarial Networks and Image Synthesis
- Advanced Image Processing Techniques
- Computer Graphics and Visualization Techniques
- Video Surveillance and Tracking Methods
- Domain Adaptation and Few-Shot Learning
- Anomaly Detection Techniques and Applications
- Human Pose and Action Recognition
- 3D Shape Modeling and Analysis
- Visual Attention and Saliency Detection
- Lung Cancer Diagnosis and Treatment
- Machine Learning and Algorithms
- Retinal Imaging and Analysis
- Privacy-Preserving Technologies in Data
- Face and Expression Recognition
- Radiomics and Machine Learning in Medical Imaging
- Adversarial Robustness in Machine Learning
- Robotics and Sensor-Based Localization
- Face recognition and analysis
- AI in cancer detection
- Video Analysis and Summarization
- Image Processing Techniques and Applications
Hong Kong Polytechnic University
2019-2025
Shanghai Jiao Tong University
2013-2025
Shenzhen Children's Hospital
2025
Shanghai University of Medicine and Health Sciences
2020-2025
Hohai University
2012-2025
Peking University
2025
Shanghai Children's Hospital
2025
Huazhong University of Science and Technology
2007-2024
Center for High Pressure Science & Technology Advanced Research
2019-2024
Fudan University Shanghai Cancer Center
2024
We show that information about social relationships can be used to improve user-level sentiment analysis. The main motivation behind our approach is users are somehow "connected" may more likely hold similar opinions; therefore, relationship complement what we extract a user's viewpoints from their utterances. Employing Twitter as source for experimental data, and working within semi-supervised framework, propose models induced either the follower/followee network or in formed by referring...
Recent transformer-based models, especially patch-based methods, have shown huge potentiality in vision tasks. However, the split fixed-size patches divide input features into same size patches, which ignores fact that elements are often various and thus may destroy semantic information. Also, vanilla transformer cannot guarantee information communication between will prevent extraction of attention with a global view. To circumvent those problems, we propose an Efficient Attention Pyramid...
In this paper, we present a method (Action-Fusion) for human action recognition from depth maps and posture data using convolutional neural networks (CNNs). Two input descriptors are used representation. The first is motion image that accumulates consecutive of action, whilst the second proposed moving joints descriptor which represents body over time. order to maximize feature extraction accurate classification, three CNN channels trained with different inputs. channel images (DMIs), both...
Domestic waste classification was incorporated into legal provisions recently in China. However, relying on manpower to detect and classify domestic is highly inefficient. To that end, this article, we propose a multimodel cascaded convolutional neural network (MCCNN) for image detection classification. MCCNN combined three subnetworks (DSSD, YOLOv4, Faster-RCNN) obtain the detections. Moreover, suppress false-positive predicts, utilized model with part judge whether results are correct....
Modeling feature interactions is of crucial significance to high-quality engineering on multifiled sparse data. At present, a series state-of-the-art methods extract cross features in rather implicit bitwise fashion and lack enough comprehensive flexible competence learning sophisticated among different fields. In this article, we propose new broad attentive graph fusion network (BaGFN) better model high-order explicit manner. On the one hand, design an module strengthen representation under...
Despite the death rate of COVID-19 is less than 3%, fatality severe/critical cases high, according to World Health Organization (WHO). Thus, screening before symptom occurs effectively saves medical resources.
Recently, machine learning models have demonstrated to be vulnerable backdoor attacks, primarily due the lack of transparency in black-box such as deep neural networks. A third-party model can poisoned that it works adequately normal conditions but behaves maliciously on samples with specific trigger patterns. However, injection function is manually defined most existing attack methods, e.g., placing a small patch pixels an image or slightly deforming before poisoning model. This results...
We introduce a new photographing guidance (PhotoHelper) for amateur photographers to enhance their portrait photo quality using deep feature retrieval and fusion. In our model, we comprehensively integrate empirical aesthetic rules, traditional machine learning algorithms neural networks extract different kinds of features in both color space aspects. With these features, build modified random forest with structured photograph collection identify types photos. also define the composition...
To conquer the bottleneck of sluggish kinetics in cathodic oxygen reduction reaction (ORR) metal-air batteries, catalysts with dual-active centers have stood out. Here, a "pre-division metal clusters" strategy is firstly conceived to fabricate N,S-dual doped honeycomb-like carbon matrix inlaid CoN4 sites and wrapped Co2 P nanoclusters as (Co2 P/CoN4 @NSC-500). A crystalline {CoII2 } coordination cluster divided by periphery second organic layers well-designed realize delocalized dispersion...
Abstract Emerging evidence suggests that modulation of gut microbiota by dietary fibre may offer solutions for metabolic disorders. In a randomized placebo-controlled crossover design trial (ChiCTR-TTRCC-13003333) in 37 participants with overweight or obesity, we test whether resistant starch (RS) as supplement influences obesity-related outcomes. Here, show RS supplementation 8 weeks can help to achieve weight loss (mean −2.8 kg) and improve insulin resistance individuals excess body...
Recently neural architecture (NAS) search has attracted great interest in academia and industry. It remains a challenging problem due to the huge space computational costs. Recent studies NAS mainly focused on usage of weight sharing train SuperNet once. However, corresponding branch each subnetwork is not guaranteed be fully trained. may only incur computation costs but also affect ranking retraining procedure. We propose multi-teacher-guided NAS, which proposes use adaptive ensemble...
A novel nonenzymatic glucose sensor was constructed based on rGO supported Pt–NiO nanoplate arrays through a simple one-step synthesis.
Abstract A total of 144 weaned piglets were used to evaluate the effects essential oil ( EO ) supplementation a low‐energy diet on performance, apparent nutrient digestibility, small intestinal morphology, microflora, immune properties and antioxidant activities in pigs. Pigs received (negative control, NC , digestible energy = 3250 kcal/kg), plus 0.025% or positive control PC 3400 kcal/kg) for 28 days. Growth performance was similar between group group. However, increased P < 0.05)...