Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks99 from public data
- 983
- MSR-GCN: Multi-Scale Residual Graph Convolution Networks for Human Motion Prediction279
A novel Multi-Scale Residual Graph Convolution Network (MSR-GCN) for human pose prediction task in the manner of end-to-end is proposed and experimental results demonstrate that the method outperforms the state-of-the-art approaches.
- From Synthetic to Real: Image Dehazing Collaborating with Unlabeled Real Data231
This work develops a disentangled image dehazing network (DID-Net), which disentangles the feature representations into three component maps, respecting the physical model of a haze process, and encourages the coarse predictions and refinements of each disentangling component to be consistent between the student and teacher networks by using a consistency loss on unlabeled real data.
- Dual Illumination Estimation for Robust Exposure Correction176
This paper presents a novel automatic exposure correction method, able to robustly produce high‐quality results for images of various exposure conditions, and demonstrates the effectiveness of the proposed approach and its superiority over the state‐of‐the‐art methods and popularautomatic exposure correction tools.
- Progressively Generating Better Initial Guesses Towards Next Stages for High-Quality Human Motion Prediction167
This paper presents a high-quality human motion pre-diction method that accurately predicts future human poses given observed ones and proposes a novel two-stage prediction frame-work, including an init-prediction network that just computes the good guess and then a formal-predictions network that predicts the target future poses based on the guess.
- 138
- 94
- Analysis of positive and negative emotions in natural scene using brain activity and GIST88
A novel emotion understanding system based on brain activity and ''GIST'' to categorize emotions reflected by natural scenes and achieve interaction with a human in terms of emotional sharing through the EEG signals is proposed.
- 75
- Rule-based extraction of experimental evidence in the biomedical domain72
The winning system is a Rule-based Information Extraction (IE) system that combines pattern matching, Natural Language Processing (NLP) tools, semantic constraints based on the domain and the specific task, and a post-processing stage for making the final curation decisionbased on the various evidence found within the document.
- Diverse Human Motion Prediction via Gumbel-Softmax Sampling from an Auxiliary Space68
This paper proposes a novel sampling strategy for sampling very diverse results from an imbalanced multimodal distribution learned by a deep generative model, which incorporates a Gumbel-Softmax coefficient matrix sampling method and an aggressive diversity promoting hinge loss function.
- Specular Highlight Removal for Real‐world Images60
This paper designs an optimization framework for simultaneously estimating the diffuse and specular highlight images from a single image and recovers the diffuse components of those regions with specular highlights by encouraging the encoding coefficients sparseness using L0 norm.
- NTIRE 2021 Challenge on Perceptual Image Quality Assessment55
The NTIRE 2021 challenge on perceptual image quality assessment (IQA), held in conjunction with the New Trends in Image Restoration and Enhancement workshop (NTIRE) workshop at CVPR 2021, achieved much better results than existing IQA methods, while the winning method can demonstrate state-of-the-art performance.
- Monte Carlo denoising via auxiliary feature guided self-attention49
A self-attention based MC denoising deep learning network based on the fact that self-Attention is essentially non-local means filtering in the embedding space which makes it inherently very suitable for theDenoising task.
- Learning to Detect Specular Highlights from Real-world Images44
This paper presents a large-scale real-world highlight dataset containing a rich variety of material categories, with diverse highlight shapes and appearances, and develops a deep learning-based specular highlight detection network (SHDNet) leveraging multi-scale context contrasted features to accurately detect specular highlights of varying scales.
- 44
- DBDNet: Learning Bi-directional Dynamics for Early Action Prediction37
A novel encoder-decoder architecture is proposed for integrating the tasks of synthesizing future motions from observed videos and reconstructing observed motions from synthesized future motions in an unified framework, which can capture the bi-directional dynamics depicted in partial videos along the temporal (past-to-future) direction and reverse chronological direction.
- 35
- 34
- 33
- Deep Camouflage Images31
A novel neural style transfer approach that adopts the visual perception mechanism to create camouflage images, which allows us to hide objects more effectively while producing natural-looking results.
- 30
- 28
- A domain-general perspective on the role of the basal ganglia in language and music: Benefits of music therapy for the treatment of aphasia28
The results show that rhythm processing plays a key role in Melodic Intonation Therapy in the rehabilitation of non-fluent aphasia patients with lesions on the basal ganglia, and provides support for taking advantage of rhythm as an important point in music therapy in clinical studies.
- Pyramid Texture Filtering26
This work shows that the approach is effective to separate structure from texture of different scales, local contrasts, and forms, without degrading structures or introducing visual artifacts, and demonstrates the applicability of the method on various applications including detail enhancement, image abstraction, HDR tone mapping, inverse halftoning, and LDR image enhancement.
- Towards High-Quality Specular Highlight Removal by Leveraging Large-Scale Synthetic Data25
A three-stage network is proposed to remove specular highlights from a single object-level image, able to generalize well to unseen real object-level images, and even produce good results for scene-level images with multiple background objects and complex lighting.
- 25
- 20
- ViDSOD-100: A New Dataset and a Baseline Model for RGB-D Video Salient Object Detection19
This work proposes a new baseline model, named attentive triple-fusion network (ATF-Net), for RGB-D video salient object detection, which aggregates the appearance information from an input RGB image, spatio-temporal information from an estimated motion map, and the geometry information from the depth map by devising three modality-specific branches and a multi-modality integration branch.
- 19
- Parsing citations in biomedical articles using conditional random fields18
The supervised machine-learning algorithms Conditional Random Fields (CRFs) are applied to automatically parse a citation into its fields (e.g., Author, Title, Journal, and Year) with a subset of html format open-access PubMed Central articles.
- 17
- 15
- RoGSplat: Learning Robust Generalizable Human Gaussian Splatting from Sparse Multi-View Images15
RoGSplat is presented, a novel approach for synthesizing high-fidelity novel views of unseen human from sparse multi-view images, while requiring no cumbersome per-subject optimization and outperforms state-of-the-art methods in novel view synthesis and cross-dataset generalization.
- A Maximum Lifetime Data Gathering Algorithm for Wireless Sensor Networks15
A near optimal maximum lifetime data gathering and aggregation algorithm MLDGA is proposed, which tries to minimize the total energy consumption in each round and maximize the lifetime of a routing tree used in the round.
- Emoji multimodal microblog sentiment analysis based on mutual attention mechanism14
A novel multimodal information integration model for emoji microblog sentiment analysis that employs a text-emoji visual mutual attention mechanism and achieves improvements in macro F1 score and accuracy.
- DNF-Intrinsic: Deterministic Noise-Free Diffusion for Indoor Inverse Rendering13
DNF-Intrinsic is presented, a robust yet efficient inverse rendering approach fine-tuned from a pre-trained diffusion model, where it is proposed to take the source image rather than Gaussian noise as input to directly predict deterministic intrinsic properties via flow matching.
- Learning to Remove Shadows from a Single Image13
Self-ShadowGAN is presented, a novel adversarial framework that is able to learn to remove shadows in an image by training solely on the image itself, using the shadow mask as the only supervision.
- 13
- Joint regression and learning from pairwise rankings for personalized image aesthetic assessment12
A novel approach for predicting personalized image aesthetics that fit an individual user’s personal taste by joint regression and learning from pairwise rankings, which is clearly outperforming state-of-the-art methods.
- 11
- 11
- BERT Based Hierarchical Sequence Classification for Context-Aware Microblog Sentiment Analysis11
This paper considers the context-aware sentiment analysis as a sequence classification task, and proposes a Bidirectional Encoder Representation from Transformers (BERT) based hierarchical sequence classification model, which extends BERT pre-trained model, with Biddirectional Long Short Term Memory (BiLSTM) and Conditional Random Field (CRF) layers.
- AEFNet: a real-time network for detecting prohibited items in x-ray images across complex scenarios10
This work proposes an adaptive efficient focusing network (AEFNet) designed to target regions, thereby enhancing the automatic detection of prohibited items and introduces the DACSP module, which dynamically adjusts sampling positions to enhance the network’s adaptability and focus on occluded targets.
- Towards High-Resolution Specular Highlight Detection10
A patch-level bidirectional refinement network for high-resolution specular highlight detection using deep learning while avoiding excessive GPU memory consumption is proposed and a modified inpainting framework for specular highlight removal as an application is proposed.
- 9
- 9
- Cross-modal and multi-level feature refinement network for RGB-D salient object detection9
A cross-modal feature interaction module is designed to enhance depth features from both channel and spatial perspectives and then effectively integrate cross- modal features and a multi-level feature fusion module is proposed which combines contextual information from multi- level features by means of skip connection.
- A blind image super-resolution network guided by kernel estimation and structural prior knowledge8
This work proposes a two stage blind super-resolution network that is based on kernel estimation strategy and is capable of integrating structural texture as prior knowledge and outperforms the state-of-the-art methods in terms of fidelity and recovery of clear details.
- Some width function asymptotics for weighted trees8
The main results include an extension of Aldous’s theorem on “search-depth” approximations by Brownian excursion to the case of weighted Galton–Watson trees, a probabilistic derivation which gen-eralizes previous results by Troutman and Karlinger on the asymptotic behavior of the expected width function and provides theuctuation law.
- NECA: Neural Customizable Human Avatar7
This work introduces NECA, an approach capable of learning versatile human representation from monocular or sparse-view videos, enabling granular customization across aspects such as pose, shadow, shape, lighting and texture.
- 7
- 7
- Reversible Data Hiding in Encrypted Images Based on Two-Round Image Interpolation6
A novel method is proposed that uses a two-round interpolation mechanism to enhance pixel prediction precision while preserving a large number of carrier pixels and can efficiently improve the capacity and fidelity compared to existing VRAE RDHEI methods.
- 5
- Transient ischemic attack analysis through non-contact approaches5
A microwave sensing platform working in C-Band and two machine learning algorithms support vector machine (SVM) and random forest (RF) are used to establish prediction models respectively to monitor transient ischemic attack in the indoor environment and improve risk management of stroke.
- 5
- 4
- Atomic filter: A weak form of shift operator for graph signals4
The concept of normal atomic filters is proposed, which degenerates into the classical shift operator under mild conditions if the graph is circulant, and which can be viewed as a weak form of the shift operator for graph signals.
- 4
- Self-supervised Texture Filtering3
This article presents a self-supervised framework for smoothing out textures while maintaining the image structures and proposes to learn texture filtering from unlabeled data by encouraging the texture inverted image generated from the filtering output to be visually more similar to the input via contrastive learning.
- 3
- When Shadow Removal Meets Intrinsic Image Decomposition: A Joint Learning Framework Using Unpaired Data3
By properly coupling shadow removal and IID into a joint learning framework, they can reinforce each other and enable promising results on both tasks, even with unpaired training data.
- 3
- 3
- Structure-Guided Diffusion Models for High-Fidelity Portrait Shadow Removal2
This work first train a shadowindependent structure extraction network on a real-world portrait dataset with various synthetic lighting conditions, which allows to generate a shadow-independent structure map including facial details while excluding the unwanted shadow boundaries.
- 2
- Multi-RoI Human Mesh Recovery with Camera Consistency and Contrastive Losses2
A multi-RoI HMR method that can estimate multiple local cameras and have the opportunity to design and apply additional constraints between cameras to improve the accuracy of the cameras and, in turn, the accuracy of the corresponding 3D mesh is proposed.
- Learning Multi‐Scale Deep Image Prior for High‐Quality Unsupervised Image Denoising2
This paper presents a novel unsupervised image denoising framework trained from single noisy image without making any explicit assumptions on the noise statistics, built upon the deep image prior (DIP), which enables diverse image restoration tasks.
- 2
- Semi-tensor product of matrices approach to stability and stabilization analysis of bounded Petri net systems2
The problem associated with marking feedback stabilizability is solved by introducing the concept and some properties of the marking pre-reachability set of BPNSs, respectively, and the necessary and sufficient condition for equilibrium point stabilization is presented.
- 2
- A key-frame-based error resilient coding scheme for video transmission over differentiated services networks2
A key-frame- based error resilient coding scheme for video transmission over differentiated services networks that can stop transmission errors propagating along the prediction loop, and can be used with on-line coding video transmission, and with precompressed video transmission.
- Face Expression Recognition via Product-Cross Dual Attention and Neutral-Aware Anchor Loss1
A neutral-expression-aware expression feature similarity loss based on the traditional anchor loss, which can further guide the network to learn better features from an input image, and which can better leverage the landmarks to extract more discriminative features from an input image.
- Incorporating Test-Time Optimization into Training with Dual Networks for Human Mesh Recovery1
This work incorporates the test-time optimization into training, performing a step of test-time optimization for each sample in the training batch before really conducting the training optimization over all the training samples, and obtains a meta-model, the meta-parameter of which is friendly to the test-time optimization.
- 1
- 1
- 1
- Enhanced Non-linear Features for On-line Handwriting Recognition Using Deep Learning1
This paper proposes and builds an on-line handwriting recognition system for French strings, which applies a DNN for non-linear feature transformation before training the character models.
- [Comparing of different methods on habitat adaptive division of Chinese material medica].1
Both model pattern and template pattern were efficient on habitat adaptive division of Chinese material medica, but they shoud bu used with different processesd and based different background.
- Generative Texture Filtering–
The core idea is to empower texture filtering by taking full advantage of the strong learned image prior of pre-trained generative models via a two-stage strategy to fine-tune a pre-trained generative model via a two-stage strategy.
- –
- –
- Make static person walk again via separating pose action from shape–
This paper proposes a method that generates pose actions in 3D space while inheriting pose shape in 2D space by separated generation of pose action and shape, and is free of entanglements of pose action and shape.
- –
- Cross-Modal Image Registration via Rasterized Parameter Prediction for Object Tracking–
This paper proposes aligning infrared and visible images using a rasterized parameter prediction algorithm with similarity measurement evaluation that achieves state-of-the-art performance on the MS-COCO and FLIR datasets.
- Corrections to: Soluble production and function of vascular endothelial growth factor/basic fibroblast growth factor complex peptide–
The State Key Laboratory of Biocontrol, School of Life Sciences, Sun Yat-Sen University, Guangzhou, China Department of Gastrointestinal Surgery, and The First Clinical School in Jinan University are investigated.
- –
- –
- –
- Distributed Optimization for Heterogenous Second-Order Multi-Agent Systems–
A continuous-time distributed optimization was researched for second-order heterogeneous multi-agent systems to make the velocities converge to the optimal value to minimize the sum of local cost functions.
- Delay Characteristic Measurement of Mobile Network at Medium Speed on Beltway–
The measurement of the delay characteristics of 4G LTE network on the beltway of Hefei City in Anhui Province of China finds that the mobile communication quality in higher speed moving scenario is not the worst when the background traffic exists during the measurement, but under the condition of lower speed the quality is relatively poor.
- Control of asynchronous sequential machine with adversarial input based on the semi-tensor product of matrices–
An algebra expression of the closed-loop system dynamic is established and the theoretical results are veri(cid:12)ed by using this algebra expression; an example is presented to illustrate the validity and application of the proposed approach.
- –
- Developing HPC Applications Based on the MIC–
This chapter gives you all the answers to how to put MIC technology into practice, and how to port the CPU programs onto the MIC platform.
- Image inpainting via variation of variances and linear weighted filling-in–
Experiments on natural images and comparisons with representative existing algorithms show that the proposed algorithm can more robustly discriminate structures and textures, estimate the best matching patch with more known information, and improve the visual quality.
- Wavelet application on abstracting stroke of Chinese character recognition–
Wavelet transformation which is a powerful tool in image analysis is used to analyze the Chinese character image in this paper, and the process of the extraction stroke is concise and distinct.
- Chinese character's branch recognition via attributed relational graph–
A new method with Neural Network on the branch matching in the recognition process of Chinese Character described by Attributed Relational Graphs is presented.
- –
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.