Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks143 from public data
- ULIP: Learning a Unified Representation of Language, Images, and Point Clouds for 3D Understanding443
ULIP is introduced to learn a unified representation of image, text, and 3D point cloud by pre-training with object triplets from the three modalities, using a pre-trained vision-language model that has already learned a common visual and textual space by training with massive image-text pairs.
- Stacked Denoise Autoencoder Based Feature Extraction and Classification for Hyperspectral Images231
Experimental results demonstrated that the SDAE pretraining in conjunction with the LR fine-tuning and classification (SDAE_LR) can achieve higher accuracies than the popular support vector machine (SVM) classifier.
- PIRM Challenge on Perceptual Image Enhancement on Smartphones: Report161
This paper reviews the first challenge on efficient perceptual image enhancement with the focus on deploying deep learning models on smartphones and proposes solutions that significantly improved baseline results defining the state-of-the-art for image enhancement on smartphones.
- LRLSHMDA: Laplacian Regularized Least Squares for Human Microbe–Disease Association prediction123
A semi-supervised computational model of Laplacian Regularized Least Squares for Human Microbe–Disease Association (LRLSHMDA) is developed by introducing Gaussian interaction profile kernel similarity calculation and LaPlacian regularized least squares classifier to achieve better prediction performance relative to the previous model.
- Open Vocabulary Object Detection with Pseudo Bounding-Box Labels122
This work proposes a method to automatically generate pseudo bounding-box annotations of diverse objects from large-scale image-caption pairs that leverages the localization ability of pre-trained vision-language models to generate pseudo Bounding- box labels and then directly uses them for training object detectors.
- RFDT: A Rotation Forest-based Predictor for Predicting Drug-Target Interactions Using Drug Structure and Protein Sequence Information112
Experimental results demonstrate that the proposed method is effective in the prediction of DTI, and can provide assistance for new drug research and development.
- 99
- Camera Placement Considering Occlusion for Robust Motion Capture82
A quality metric that estimates the error caused by both image resolution and occlusion is proposed that reflects the dynamic self-occlusion of the target and can be applied toward the automatic design of more accurate and robust tracking systems.
- 79
- Range Scaling Global U-Net for Perceptual Image Enhancement on Mobile Devices54
This paper presents a novel deep learning based approach—the Range Scaling Global U-Net (RSGUNet)—for perceptual image enhancement on mobile devices that learns a global feature vector as well as a novel range scaling layer that alleviate artifacts in the enhanced images.
- An occlusion metric for selecting robust camera configurations38
This work introduces a configuration quality metric based on the likelihood of dynamic occlusion, which is extensively evaluated experimentally using hundreds of different camera configurations and found to correlate very closely with the actual probability of feature occlusions.
- GlueGen: Plug and Play Multi-modal Encoders for X-to-image Generation33
GlueGen is proposed, which applies a newly proposed GlueNet model to align features from single-modal or multi-modal encoders with the latent space of an existing T2I model, allowing for flexible and efficient integration of new functionality into existing T2I models and sheds light on X-to-image (X2I) generation.
- Mask-Free OVIS: Open-Vocabulary Instance Segmentation without Manual Mask Annotations27
This work learns both base and novel categories from pseudo-mask annotations generated by the vision-language model in a weakly supervised manner using the proposed Mask-free OVIS pipeline, which significantly improves the mAP scores on the MS-COCO dataset and OpenImages dataset.
- Identification of self-interacting proteins by exploring evolutionary information embedded in PSI-BLAST-constructed position specific scoring matrix25
A novel computational approach called RVMBIGP that combines the Relevance Vector Machine (RVM) model and Bi-gram probability (BIGP) to predict SIPs based on protein sequence is proposed and may provide a cost-effective alternative for Sips identification.
- 24
- 23
- Distributed image understanding with semantic dictionary and semantic expansion22
This paper proposes a semantic dictionary to describe the images on the level of semantic, which characterizes the probability distribution between visual appearances and semantic concepts, and the learning procedure of semantic dictionary is formulated into a minimization optimization problem.
- On Nonparametric Image Registration22
This article suggests solving the IR problem locally, by first studying the local properties of a mapping transformation, and proposes a local smoothing method that does not require any parametric form or other global regularity conditions on the mapping transformation.
- Total restrained domination in graphs22
The best possible upper and lower bounds for @c"t"r(G) are determined, those graphs achieving these bounds are characterized and the best possible lower bounds are found.
- Feature based image registration using non-degenerate pixels21
A new type of image features is studied, which has the following two properties: they should be easy to identify by a computer algorithm so that the entire feature extraction procedure is computer automatic, and they should provide proper information to approximate the geometric matching transformation accurately.
- 20
- Moving Target Shadow Analysis and Detection for ViSAR Imagery18
The frame rate and the shadow feature are analyzed, the velocity limitation of ViSAR moving target shadow detection is discussed and the expression of velocity limitation is quantitatively given.
- 17
- 17
- Spectral 3D reconstruction of impressionist oil paintings based on macroscopic OCT imaging15
This work presents a hybrid scanning platform combined with an effective algorithm for real-time sampling and artifact removal to achieve macroscopic OCT (macro-OCT) imaging and spectral 3D reconstruction of impressionist style oil paintings.
- A Visual and Semantic Image Retrieval Method Based on Similarity Computing with Query-Context Recognition14
An image retrieval method based on visual and semantic similarity computing with a query-context recognition mechanism and it is found that users' query intention can be detected from the difference of images in queries.
- 13
- An Image-Query Creation Method for Representing Impression by Color-based Combination of Multiple Images12
This image-query creation method representing impression of color makes it possible to expand the range of image retrieval for a large number of image data of cultural property in digital archives, such as electronic library and electronic museum, automatically.
- CSLSEP: an ensemble pruning algorithm based on clustering soft label and sorting for facial expression recognition11
A novel ensemble pruning algorithm based on clustering soft label optimization and sorting for facial expression recognition that outperforms or competes with some state-of-the-art methods on several typical facial expression datasets.
- HTSTL: Head-and-Tail Search Network With Scale-Transfer Layer for Traffic Sign Text Detection10
An effective detector for traffic sign text is designed, whose pipeline only consists of a preprocessing module to tackle with some complex situations, a Fully Convolutional Network (FCN) in which a Scale-transfer layer is proposed to speed up the network and a simple post-processing step.
- An Improved Generalized Chirp Scaling Algorithm Based on Lagrange Inversion Theorem for High-Resolution Low Frequency Synthetic Aperture Radar Imaging10
An improved GCSA based on Lagrange inversion theorem is analytically derived and enables the high-order range-dependent coupling phase to be accurately compensated and the image quality and focusing depth in range dimension are greatly improved.
- 10
- Predictive Model for Concentration Distribution of Explosive Dispersal9
This work derives a model to predict the concentration distribution of liquid and granular material dispersal, considering the two questions above, and was validated using experimental data on the mean concentration of dispersal and was well fitted.
- A Combined Image-Query Creation Method for Expressing User's Intentions with Shape and Color Features in Multiple Digital Images9
This method uses image databases provided for query-creation and performs several set-operators to express user's imagination by combining user's imaginary images and real scenes to create an imaginary image as the combined-image query.
- 8
- 8
- Elevation Spatial Variation Analysis and Compensation in GEO SAR Imaging8
An imaging algorithm is proposed to solve the 3D space variation and improve the focusing depth in elevation and simulations with dot-matrix targets and distributed targets are performed to validate the imaging method.
- Diagnosis model of pancreatic cancer based on fusion of distribution estimation algorithm and genetic algorithm8
A pancreatic cancer diagnosis model based on the fusion of distribution estimation algorithm and genetic algorithm is proposed, which shows that the diagnostic model constructed has a very high application value in the early auxiliary pre-diagnosis of pancreatic cancer.
- 7
- 7
- An ensemble pruning method considering classifiers’ interaction based on information theory for facial expression recognition6
This paper proposes a new ensemble pruning method CCIEP (An Ensemble Pruning Method Considering Classifiers' Interaction), which outperforms some state-of-the-art ensemble pruning methods on five classical face expression data sets and 10 UCI data sets.
- 6
- 6
- 5
- 5
- Deriving Semantic from Images Based on the Edge Information.5
Most extracted independent factors based on the method are demonstrated to be related to human semantic according to the experiments which are performed on 7,000 images.
- Pairwise dependency-based robust ensemble pruning for facial expression recognition4
A novel algorithm called Pairwise Dependency-based Robust Ensemble Pruning (PDREP) is proposed, which treats the predicted results of classifiers for sample instances as features of the classifier and evaluates their dependencies between pairs of classifiers using mutual information to perform robust classifier pruning.
- 4
- 4
- Research on defect inspection method of pipeline robot based on adaptive image enhancement4
The experimental results show that the detection method in this paper can adaptively correct the image brightness, and the uneven brightness of the image is significantly improved.
- 4
- 4
- Driver Fatigue Detection Based on SVM4
A method to detect driver fatigue with support vector machine is introduced and results show that fatigue detection accuracy rate reached 97.80%, and the method is suitable for driver fatigue detection.
- Intensity based nonparametric image registration4
An intensity-based image registration method without requiring a parametric form of the mapping transformation, which can be nonparametric, and it can even be discontinuous at certain places in the design space is proposed.
- A framework for a feedback process to analyze and personalize a document vector space in a feature extraction model4
A framework for a feedback process to implement a highly accurate document retrieval system and several experimental results are presented to clarify how retrieval accuracy was improved by the feedback process and how accurately documents that satisfied the purpose and interests of users were extracted.
- 3
- 3
- 3
- 3
- 3
- 3
- 3
- 3
- 3
- 3
- 3
- A Personalized Recommender System Model Using Colour-impression-based Image Retrieval and Ranking Method3
This paper presents an approach for extraction of colour-properties of images in order to represent impressions related to the human perception of images and demonstrates how this can be accomplished by a personalized recommender system model that is based on analysis of colour features of product images.
- 3
- 2
- 2
- Profile-Guided Parallel Task Extraction and Execution for Domain Specific Heterogeneous SoC2
The integrated approach offers a path forward for application developers to take full advantage of the target SoC without requiring users to become hardware and parallel programming experts.
- Intelligent Active and Passive Learning for Integrated Semantic Computing for Vision Data Annotation2
- Vector Quantization: Timeline-Based Location Data Extraction and Route Fitting for Crowdsourcing2
The idea of Vector Quantization, denoted as VQ, is referred to, and a model of TDERTM (Timeline-based Data Extraction and Route Fitting Model) is given to solve trajectory data extraction of timeline-based location information and route fitting in the field of military and civilian research.
- Cross-Layer Convolutional Siamese Network for Visual Tracking2
A cross-layer convolutional Siamese network tracker which attempts to explore more semantic features of different layers from two aspects, and demonstrates the effectiveness of Siam-CC as compared to the state-of-the-art trackers.
- Direct Torque Control Servo System for Permanent Magnet Synchronous Motor2
The hardware and software of PMSM servo system based on the advance DSP of TMS302F28335 were designed and the principle of DTC for PMSM and stator flux estimator used LPF instead of pure integrator are introduced.
- 2
- 1
- 1
- 1
- Design and implementation of indoor robot positioning and mapping based on ROS1
A mobile robot system based on ROS operating system and Lidar and the STM32 single-chip microcomputer is designed and constructed to solve the mapping and positioning requirements of indoor robots.
- 1
- An Artificial Intelligence and Internet of Things Platform for Healthcare and Industrial Applications1
This chapter presents an artificial intelligence (AI) and Internet of Things (IoT) platform that shares many supporting functions including the web/mobile app API, user management, and device management.
- 1
- Fatigue detection based on multi-feature fusion of fatigue behavior1
The experimental results indicate that the proposed method could accurately and real-time detect whether the tester is in a state of fatigue driving, and also determine the fatigue level.
- 1
- 1
- 1
- 1
- 1
- A light-weight feedback method for reconstructing a document vector space on a feature extraction model1
A document retrieval system with a light-weight feedback method for reconstructing a document vector space, developed on a Feature Extraction Model (FEM), which can apply the feedback method dynamically to reconstruct the vector spaces based on intensions of users.
- 1
- 1
- 1
- –
- –
- –
- Proteomics for Biomarker Discovery in Gynecological Cancers: A Systematic Review–
Collagens, fibrinogens, chaperones, and apolipoproteins were revealed to be replicated in GCs and to be regulated consistently.
- –
- –
- –
- Research on saliency detection algorithm of rural homestead–
Four commonly used algorithms of FT, HC, LC and AC with good saliency detection effect are studied, showing that the LC algorithm is more suitable for theSaliency detection of rural homestead images in this paper.
- Micro-motion detection based on pulsed coherent radar (PCR) in high reflection situations–
This paper focuses on the analysis of Acconeer's PCR10P development board human presence detection for high inverse static object false alarm and proposes a high inverse suppression algorithm based on the target phase.
- Fatigue detection based on non-contact respiratory detection–
A large number of experimental comparisons have confirmed that fatigue can be detected faster through respiratory rate fatigue characteristics, but there will still be a small deviation in the detection of respiratory rate, so this paper proposes to compare facial fatigue characteristics with breathing.
- –
- –
- –
- –
- –
- –
- Building Contour Detection Based on Straightness Level Set–
An improved level set algorithm is proposed, which combined the traditional level set and chain code, then it introduced the parameters of straightness, and could automatically detect whether the segmentation curve is a building by calculating the straightness.
- –
- –
- UAV Remote Sensing Image Rural Building Detection Based on Convolutional Neural Network–
This paper uses massive remote sensing data of UAV to train convolutional neural network and improves CaffeNet model and the improved a-CaffeNet has an accuracy rate of 97.30%.
- –
- –
- –
- –
- –
- –
- –
- –
- –
- Color block-histogram image retrieval based on difference degree of spatial distribution–
The experimental results show that this algorithm can overcome the disadvantages of the traditional method, and has better retrieval performance.
- An Adaptive Watermark Authentication System Based on Image Content Features–
An adaptive image watermarking authentication system that extracts image content features as watermark information, and uses human visual system characteristics to realize embedding self-adaptive and blind detection function for the water marking.
- –
- –
- –
- Image registration by non-degenerate pixel detection–
Solving the image registration problem locally, by first studying the local properties of a mapping transformation, is suggested, and an intensity-based local smoothing method for estimating the geometrical transformation is proposed.
- –
- Video Signal Capture and Processing Card Based on H.264 Hardware Encoder–
This paper proposes a hardware and software design proposal of video collection and processing used in digital multimedia microteaching system that adopts friendly operating interface, the user operation is very convenience.
- –
- –
- –
- –
- –
- –
- –
- –
- Processing of Disk Cam Based on User’s Macro-program–
The general manual programming and computer aided programming are not good for the manufacture of cam and the macro-programming can simply and effectively process the cam.
- –
- –
- –
- –
Publication data from OpenAlex, with missing venues and authors filled in from Crossref; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-15. One-sentence summaries under some papers are written by Semantic Scholar’s model. Citation counts may be lower than on Google Scholar, which indexes more sources.
Report an error
Wrong papers, two people merged into one, or a profile that should not be here? Tell us and we will fix or hide it.