A5019433605- Advanced Image and Video Retrieval Techniques
- Advanced Vision and Imaging
- Image Enhancement Techniques
- Image Retrieval and Classification Techniques
- Video Surveillance and Tracking Methods
- Multimodal Machine Learning Applications
- Medical Image Segmentation Techniques
- Anomaly Detection Techniques and Applications
- Advanced Radiotherapy Techniques
- Computer Graphics and Visualization Techniques
- Advanced Image Processing Techniques
- Generative Adversarial Networks and Image Synthesis
- Robotics and Sensor-Based Localization
- Human Pose and Action Recognition
- Domain Adaptation and Few-Shot Learning
- Advanced Neural Network Applications
- Image and Signal Denoising Methods
- Network Security and Intrusion Detection
- Face recognition and analysis
- Face and Expression Recognition
- Video Analysis and Summarization
- Underwater Vehicles and Communication Systems
- Visual Attention and Saliency Detection
- Radiation Therapy and Dosimetry
- Radiation Dose and Imaging
Shenzhen Institutes of Advanced Technology
2014-2025
Chinese Academy of Sciences
2014-2024
Huazhong University of Science and Technology
2024
Tongji Hospital
2024
Shenzhen University
2013-2024
University of Maryland, Baltimore
2017-2024
University of Science and Technology Beijing
2024
Xinjiang Normal University
2023
University of Chinese Academy of Sciences
2022-2023
Xinjiang Institute of Ecology and Geography
2021-2022
Recently, convolutional neural networks (CNNs) have achieved great improvements in single image dehazing and attained much attention research. Most existing learning-based methods are not fully end-to-end, which still follow the traditional procedure: first estimate medium transmission atmospheric light, then recover haze-free based on scattering model. However, practice, due to lack of priors constraints, it is hard precisely these intermediate parameters. Inaccurate estimation further...
Latest diffusion-based methods for many image restoration tasks outperform traditional models, but they encounter the long-time inference problem. To tackle it, this paper proposes a Wavelet-Based Diffusion Model (WaveDM). WaveDM learns distribution of clean images in wavelet domain conditioned on spectrum degraded after transform, which is more time-saving each step sampling than modeling spatial domain. ensure performance, unique training strategy proposed where low-frequency and...
Denoising diffusion models have emerged as a powerful tool for various image generation and editing tasks, facilitating the synthesis of visual content in an unconditional or input-conditional manner. The core idea behind them is learning to reverse process gradually adding noise images, allowing generate high-quality samples from complex distribution. In this survey, we provide exhaustive overview existing methods using editing, covering both theoretical practical aspects field. We delve...
Video Recognition has drawn great research interest and progress been made. A suitable frame sampling strategy can improve the accuracy efficiency of recognition. However, mainstream solutions generally adopt hand-crafted strategies for It could degrade performance, especially in untrimmed videos, due to variation frame-level saliency. To this end, we concentrate on improving video classification via developing a learning-based strategy. We intuitively formulate procedure as multiple...
Computed Tomography (CT) plays an important role in monitoring radiation-induced Pulmonary Fibrosis (PF), where accurate segmentation of the PF lesions is highly desired for diagnosis and treatment follow-up. However, task challenged by ambiguous boundary, irregular shape, various position size lesions, as well difficulty acquiring a large set annotated volumetric images training. To overcome these problems, we propose novel convolutional neural network called PF-Net incorporate it into...
Lithology identification is of great importance in reservoir characterization. Recently, many researchers have applied machine-learning techniques to solve lithology problems from well-log curves, and their works indicate three main characteristics. First, most predict lithofacies using features measured during logging, whereas very few consider adding stratigraphic sequence information that available prior drilling this problem. Second, studies properties one depth point, take the influence...
Vegetation change and ecological quality of the Loess Plateau (LP) are directly related to protection high-quality development Yellow River Basin. Based on LP zoning multisource remote sensing data, we analyzed vegetation its relationship with climate, terrestrial water storage (TWS), land use/cover from 2000 2020, using normalized difference index (NDVI), fraction cover (FVC), net primary productivity (NPP). And environmental was evaluated based (RSEI). The results showed that spatial...
Taitema Lake, located in the lower reaches of Tarim River and Cherchen River, is one most important ecological barriers Ruoqiang County. The amount water Lake plays an role maintaining a healthy cycle within ecosystem, curbing sandstorms, improving salinization desertification. aim this study was to reasonably determine volume conveyance by calculating demand. We systematically analyzed spatial temporal variation characteristics during 21 processes from 2000 2020. results showed that area...
The popularization of the internet and widespread use smartphones have led to a rapid growth in number social media users. While information technology has brought convenience people, it also given rise cyberbullying, which serious negative impact. identity online users is hidden, due lack supervision imperfections relevant laws policies, cyberbullying occurs from time time, bringing mental harm psychological trauma victims. pre-trained language model BERT (Bidirectional Encoder...
Denoising diffusion models have emerged as a powerful tool for various image generation and editing tasks, facilitating the synthesis of visual content in an unconditional or input-conditional manner. The core idea behind them is learning to reverse process gradually adding noise images, allowing generate high-quality samples from complex distribution. In this survey, we provide exhaustive overview existing methods using editing, covering both theoretical practical aspects field. We delve...
This paper aims to introduce the robustness against noise into spectral clustering algorithm. First, we propose a warping model map data new space on basis of regularization. During warping, each point spreads smoothly its spatial information other points. After empirical studies show that clusters become relatively compact and well separated, including cluster is formed by In this space, number can be estimated eigenvalue analysis. We further apply mapping obtain low-dimensional...
Image segmentation plays an important role in computer vision and image analysis. In this paper, is formulated as a labeling problem under probability maximization framework. To estimate the label configuration, iterative optimization scheme proposed to alternately carry out maximum posteriori (MAP) estimation likelihood (ML) estimation. The MAP modeled with Markov random fields (MRFs) graph cut algorithm used find solution ML achieved by computing means of region features Gaussian model....
Convolutional neural networks (CNNs) have attracted much research attention and achieved great improvements in single-image dehazing. However, previous learning-based dehazing methods are mainly trained on synthetic data, which greatly degrades their generalization capability natural hazy images. To address this issue, article proposes a semi-supervised learning approach for dehazing, where both realistic images leveraged during training. Considering the situation that it is hard to obtain...
Large language models (LLMs), such as GPT-4, have shown remarkable performance in natural processing (NLP) tasks, including challenging mathematical reasoning. However, most existing open-source are only pre-trained on large-scale internet data and without math-related optimization. In this paper, we present WizardMath, which enhances the reasoning abilities of Llama-2, by applying our proposed Reinforcement Learning from Evol-Instruct Feedback (RLEIF) method to domain math. Through...
Neural radiance fields have made a remarkable breakthrough in the novel view synthesis task at 3D static scene. However, for 4D circumstance (e.g., dynamic scene), performance of existing method is still limited by capacity neural network, typically multilayer perceptron network (MLP). In this article, we utilize Voxel to model field, short as V4D, where voxel has two formats. The first one regularly space and then use sampled local feature with time index density field texture tiny MLP....
Though action recognition in videos has achieved great success recently, it remains a challenging task due to the massive computational cost. Designing lightweight networks is possible solution, but may degrade performance. In this paper, we innovatively propose general dynamic inference idea improve efficiency by leveraging variation distinguishability of different videos. The approach can be from aspects network depth and number input video frames, or even joint input-wise depth-wise...
Deep supervised hashing has emerged as an effective solution to large-scale semantic image retrieval problems in computer vision. Convolutional neural network-based methods typically seek pairwise or triplet labels conduct similarity-preserving learning. However, complex concepts of visual contents are hard capture by similar/dissimilar labels, which limits the performance. Generally, losses not only suffer from expensive training costs but also lack sufficient information. In this paper, we...
In this paper, a computation efficient regression framework is presented for estimating the 6D pose of rigid objects from single RGB-D image, which applicable to handling symmetric objects. This designed in simple architecture that efficiently extracts point-wise features data using fully convolutional network, called XYZNet, and directly regresses without any post refinement. case object, one object has multiple ground-truth poses, one-to-many relationship may lead estimation ambiguity....
Abstract Background Necrotizing enterocolitis (NEC) is a serious gastrointestinal disease, primarily affects preterm newborns and occurs after 7 days of life (late-onset NEC, LO-NEC). Unfortunately, over the past several decades, not much progress has been made in its treatment or prevention. This study aimed to analyze risk factors for LO-NEC, impact LO-NEC on short-term outcomes very infants (VPIs) with focus nutrition different onset times. Method Clinical data VPIs were retrospectively...
(a) To investigate the accuracy of cone-beam computed tomography (CBCT)-derived dose distributions relative to fanbeam-based simulation CT-derived distributions; and (b) study feasibility CBCT dosimetry for guiding appropriateness replanning.Image data corresponding 40 patients (10 head neck [HN], 10 lung, pancreas, pelvis) who underwent radiation therapy were randomly selected. Each patient had both intensity-modulated volumetric-modulated arc plans; these 80 plans subsequently recomputed...