A5101588964- Cryptographic Implementations and Security
- Advanced Vision and Imaging
- Advanced Electron Microscopy Techniques and Applications
- Privacy-Preserving Technologies in Data
- Advanced Malware Detection Techniques
- Optical measurement and interference techniques
- Color Science and Applications
- Integrated Circuits and Semiconductor Failure Analysis
- Image Enhancement Techniques
- Advances in Cucurbitaceae Research
- Photovoltaic System Optimization Techniques
- Impact of Light on Environment and Health
- Electron and X-Ray Spectroscopy Techniques
- 3D Shape Modeling and Analysis
- Chaos-based Image/Signal Encryption
- Advanced Sensor and Control Systems
- Image Processing Techniques and Applications
- Nanoplatforms for cancer theranostics
- Stochastic Gradient Optimization Techniques
- Bluetooth and Wireless Communication Technologies
- Visual Attention and Saliency Detection
- Industrial Technology and Control Systems
- Image and Video Quality Assessment
- Visual perception and processing mechanisms
- Optical Imaging and Spectroscopy Techniques
Shanghai Jiao Tong University
2011-2025
Shanghai Sixth People's Hospital
2022-2025
Zhejiang University
2021-2024
State Key Laboratory of Cryptology
2024
Westlake University
2024
Fudan University
2016-2023
Tencent (China)
2023
Collaborative Innovation Center of Chemistry for Energy Materials
2023
University of Science and Technology of China
2021
Institute of Computing Technology
2013-2021
Osteoporosis is a chronic condition affecting patients' morbidity and mortality represents big socioeconomic burden. Because stem cells can proliferate differentiate into bone-forming cells, cell therapy for osteoporosis has been widely studied. However, as live drug face multiple challenges because of their instability during preservation transportation. In addition, potential adverse effects such embolism, tumorigenicity, immunogenicity.Herein, we sought to use cell-mimicking targeted...
Reconstructing accurate 3D scenes from images is a long-standing vision task. Due to the ill-posedness of single-image reconstruction problem, most well-established methods are built upon multi-view geometry. State-of-the-art (SOTA) monocular metric depth estimation can only handle single camera model and unable perform mixed-data training due ambiguity. Meanwhile, SOTA trained on large mixed datasets achieve zero-shot generalization by learning affine-invariant depths, which cannot recover...
Fully exposed cluster catalysts (FECCs) can not only provide the catalytic sites with multiple metal atoms as in nanoparticles (NPs) but also maintain a full atomic utilization efficiency single-atom (SACs), making them bridge linking SACs and NPs. Due to these particular characteristics, FECCs have been widely studied exhibited enhanced performance dehydrogenation oxidation reactions. However, their application hydrogenation was rarely reported, reaction mechanism has yet be understood....
We introduce Metric3D v2, a geometric foundation model designed for zero-shot metric depth and surface normal estimation from single images, critical accurate 3D recovery. Depth estimation, though complementary, present distinct challenges. State-of-the-art monocular methods achieve generalization through affine-invariant depths, but fail to recover real-world scale. Conversely, current techniques struggle with performance due insufficient labeled data. propose targeted solutions both...
Significant strides have been made in the development of cancer vaccines to combat malignant tumors. However, natural immunosuppressive environment within tumors, known as tumor microenvironment (TME), hampers uptake and presentation antigens by antigen-presenting cells (APCs) itself. This limitation results inadequate activation immune responses against cancer. In contrast, peritumoral tissue maintain their normal functions. this context, we present an interesting approach enhance...
Deep learning (DL)-based techniques have recently proven to be very successful when applied profiled side-channel attacks (SCA). In a real-world SCA scenario, attackers gain knowledge about the target device by getting access similar prior attack. However, most state-of-the-art literature performs only proof-of-concept attacks, where traces intended for profiling and attacking are acquired consecutively on same fully-controlled device. This paper reminds that even small discrepancy between...
Cognitive ability evaluation in intelligent vehicles is conventionally evaluated by classical autonomous driving dataset, which lacks comprehensive annotations of difficulty. Realistically, different conditions require vast level cognitive ability, e.g., highly congested traffic much more challenging than on limited access highway; a blizzard/hurricane requires robust environmental cognition abilities under ordinary conditions. Different datasets contain proportions various conditions,...
Single-particle cryo-electron microscopy (cryoEM) methods are powerful for solving high-resolution structures of biological macromolecules. Locating numerous particles from micrographs is essential three-dimensional reconstruction but challenging due to the extremely low signal-to-noise ratio and various particle shapes in micrographs. In this study, we devise TransPicker, a two-dimensional picking framework based on novel end-to-end transformer-based detective method named crDETR (cryoEM...
Although dimming the light in tunnels brings salient benefits to energy saving, effects of dimming-induced luminance changes on driving safety have been rarely explored. Adopting visual performance experiment, present study investigated impacts change rate (V) driver's perception, judgment, and operation threshold zone tunnel under different seasons weather conditions. The results show that reaction times (RTs), pupil area (vp), blink frequency (fb), increased with increase V. When before...
Mutation of the isocitrate dehydrogenase (IDH) gene is regarded a novel indicator for prognosis patients with glioma. However, role IDH1 mutations in carcinogenesis and mechanisms underlying their function glioblastoma multiforme (GBM) remain unknown. The present study aimed to determine whether association RLIP76 different mutational status could serve as putative biomarker improving disease prognosis. Quantitative PCR, western blotting immunohistochemical staining assays were used...
For grid-connected converters (GCC), the synchronization unit (SU) is indispensable for vector current control (VCC). Usually, phase-locked loop (PLL) employed as SU. However, PLL could introduce negative resistance contributing to potential instability issues. Thus, second-order generalized integrator (SOGI-PLL) and reactive power-based method (RPS) are proposed enhance stability. Nevertheless, intrinsic reason why RPS enhances stability not clear. Therefore, a detailed comparative analysis...
In this study, we address the challenge of 3D scene structure recovery from monocular depth estimation. While traditional estimation methods leverage labeled datasets to directly predict absolute depth, recent advancements advocate for mix-dataset training, enhancing generalization across diverse scenes. However, such mixed dataset training yields predictions only up an unknown scale and shift, hindering accurate reconstructions. Existing solutions necessitate extra or geometry-complete...
Profiled attacks play a fundamental role in evaluating the security of cryptographic algorithm implementations worst case. In recent years, deep learning has been introduced as new powerful alternative to achieve these attacks. Many works have demonstrated that could solve desynchronization issue and break masking countermeasures. From early years when researchers explore machine methods, there is road map profiled be converted non-profiled ones by on-the-fly strategy. TCHES 2019,...
Cryo-electron microscopy (cryo-EM) single particle analysis (SPA) has been an indispensable technology to reconstruct three-dimensional (3D) structures of biomolecules at near-atomic resolution. Tens thousands particles are required obtain high-resolution 3D reconstructions, nevertheless, it is rather challenging due the extremely noisy images and diversity particles. Recently, while deep learning-based methods have devoted into improvement feature extraction location estimation, most them...