Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks76 from public data
- Implicit Dual-Domain Convolutional Network for Robust Color Image Compression Artifact Reduction106
An implicit dual-domain convolutional network with a pixel position labeling map and quantization tables as inputs is proposed and is superior to the state-of-the-art methods and IDCN-f exhibits excellent abilities to handle a wide range of compression qualities with a little trade-off against performance.
- 87
- AIM 2019 Challenge on Image Demoireing: Methods and Results57
This paper reviews the first-ever image demoireing challenge that was part of the Advances in Image Manipulation (AIM) workshop, held in conjunction with ICCV 2019, and focuses on the proposed solutions and their results.
- Pneumonia Detection Using an Improved Algorithm Based on Faster R-CNN49
An improved algorithm that can provide doctors with location information of pneumonia lesions is proposed and obtained 39.23% Mean Average Precision on the X-ray image dataset from the Radiological Society of North America (RSNA) and got 38.02% mean average precision on the ChestX-ray14 dataset, surpassing other detection algorithms.
- A Cascade Face Spoofing Detector Based on Face Anti-Spoofing R-CNN and Improved Retinex LBP48
This paper designs face anti-spoofing region-based convolutional neural network (FARCNN), based on improved Faster region- based Convolutional Neural Network (R-CNN) framework, which is motivated by face detection and regards the face spoofing detection as a three-way classification to distinguish real face, fake face and background.
- 41
- 36
- An Improved Algorithm for Detecting Pneumonia Based on YOLOv335
This study proposes an improved algorithm that can provide doctors with location information on lesions for the detection of pneumonia and obtained 46.84 mean average precision on the X-ray image dataset provided by the Radiological Society of North America (RSNA), surpassing other detection algorithms.
- 35
- Field-programmable gate array implementation of a probabilistic neural network for motor cortical decoding in rats35
A field-programmable gate array (FPGA) implementation of a probabilistic neural network (PNN) is proposed and developed to decode motor cortical ensemble recordings in rats performing a lever-pressing task for water rewards, indicating that the performance of current FPGAs is competent for portable BMI applications.
- 32
- 31
- 31
- 28
- S-Net: a scalable convolutional neural network for JPEG compression artifact reduction25
Experimental results indicate that the proposed S-Net CNN outperforms other CNN-based methods and achieves state-of-the-art performance.
- 24
- 23
- 22
- Fast transcoding from H.264 to HEVC based on region feature analysis21
A fast H.264 to HEVC transcoding algorithm based on region feature analysis achieves a significant improvement on transcoding speed, while maintaining high Rate-Distortion performance.
- Dynamic DNN Decomposition for Lossless Synergistic Inference19
D, a dynamic DNN decomposition system for synergistic inference without precision loss is proposed, which introduces a heuristic algorithm named horizontal partition algorithm to split a DNN into three parts and can partially adjust the partitions at run time according to processing time and network conditions.
- GeminiNet: Combine Fully Convolution Network With Structure of Receptive Fields for Object Detection15
This article designs GeminiNet to identify and localize the pneumonia in Chest X-ray (CXR) images using a popular fully convolution architecture with computation shared on the entire image, combining RoI Align and PSRoI Pooling to capture global and local information and output.
- Robust visual tracking for planar objects using gradient orientation pyramid15
A robust DVT method, named gradient orientation pyramid efficient second-order minimization (GOP-ESM), based on the proposed gradient orientation Pyramid descriptor is presented, which outperforms the state-of-the-art tracking methods against various environmental variations, especially illumination changes.
- 13
- 11
- 10
- 10
- 8
- Image Denoising via Multi-Scale Gated Fusion Network7
A multi-scale gated fusion network (MGFN) for image denoising, which learns direct end-to-end mappings from corrupted images to clean images and can well recover textures, yielding favorable performance against other state-of-the-art methods.
- 7
- A video structural similarity quality metric based on a joint spatial-temporal visual attention model7
A structural similarity quality metric for videos based on a spatial-temporal visual attention model that takes the burst of error into account by introducing the non-linear weighting functions to give a much higher weighting factor to the extremely damaged frames.
- 7
- Spatial likelihood voting with self-knowledge distillation for weakly supervised object detection6
In this framework, a spatial likelihood voting (SLV) module is introduced to converge region proposal localization without bounding box annotations and a self-knowledge distillation module is proposed to refine the feature representations of the given image.
- Adaptive coding-unit size selection based on hierarchical quad-tree correlations for high-efficiency video coding6
A fast CU size selection algorithm based on hierarchical quad-tree correlations (HQTCs) is proposed that can achieve on average a 26% computational time reduction under all configurations with a negligible BD-rate increase of 0.47% compared with the original encoding scheme in HEVC test model HM13.0.
- 5
- Learning Robust Features for Planar Object Tracking5
A novel network architecture is employed, the efficient second-order minimization (ESM) layer, which performs the ESM algorithm on deep feature maps for improving the robustness of direct methods in visual tracking.
- 5
- 5
- 5
- Hard parameter sharing for compressing dense-connection-based image restoration network4
The distribution of convolution layers is studied and a hard parameter sharing approach known as convolution pool (CP) is proposed for compressing dense-connection-based image restoration CNN models, which can efficiently achieve an impressive compression rate with negligible performance reduction.
- Video super-resolution with phase-aided deformable alignment network4
In PDAN, a phase-based method is introduced in the motion estimation subnetwork along with the traditional U-Net structure, which improves the restoration robustness in scenarios of lighting change and a deformable convolutional network is adopted in the alignment subnetwork to enhance the alignment for objects with an irregular shape and motion distortion.
- Convolutional neural network with uncertainty estimates for no-reference image quality assessment4
This paper proposes a CNN with two branches for general noreference image quality assessment, one branch of this model predicts the patch quality, and the other predicts the uncertainty, which denotes the degree of deviation of the patchquality from the image quality.
- 4
- Phase-based frame rate up-conversion for depth video4
A phase-based method to interpolate depth frames without explicitly finding correspondences is proposed, combining top-down phase difference preservation and bottom-up phase difference correction.
- Hole-filling map-based coding unit size decision for dependent views in three-dimensional high-efficiency video coding4
A fast coding unit (CU) size decision algorithm for both depth and texture components in dependent views, where hole-filling maps created through view synthesis are utilized, which can achieve on average 54.3% time reduction.
- Multi-Label Zero-Shot Learning Via Contrastive Label-Based Attention3
This study proposes a contrastive label-based attention method (CLA) to associate each label with the most relevant image regions, guided by the latent label embedding, which outperforms the state-of-the-art methods.
- A Unified Asymmetric Knowledge Distillation Framework for Image Classification3
Experimental results show that the proposed concise distillation framework can achieve considerable improvements over existing methods, demonstrating the effectiveness of the method and the potential of the framework.
- 3
- 3
- 3
- 3
- 2
- Build receptive pyramid for efficient color image compression artifact reduction2
This work presents an RP-based convolution neural network named receptive pyramid convolutional network (RPCN) for efficient color image compression artifact reduction (CAR) with a much smaller model and much faster running speed for the CAR task.
- 2
- Deep supervised dictionary learning for no-reference image quality assessment2
A deep convolutional neural network for general no-reference image quality assessment (NR-IQA), i.e., accurate prediction of image quality without a reference image, is proposed and its performance is comparable with that of state-of-the-art general-purpose NR- IQA algorithms.
- Multitask convolutional neural network for no-reference image quality assessment2
A multitask convolutional neural network for general no-reference image quality assessment (NR-IQA) that decomposes the task of rating image quality into two subtasks, namely distortion identification and distortion-level estimation, and then combines the results of the two subt tasks to obtain a final image quality score.
- Content based error resilience coding using redundant coding, reference frame modification, and mode decision2
Experimental results demonstrate that the proposed algorithm achieves significant performance gains over the LA-RDO and HRP method when video is transmitted over error-prone channel.
- A Fast Intra Coding Unit Size Decision Based on Statistical Learning for HEVC2
A fast intra coding unit size decision algorithm is proposed to reduce the heavy complexity of HEVC encoding and reduces unit sizes search by using the classifier, which is based on the statistical learning.
- Video image assessment with a distortion-weighing spatiotemporal visual attention model2
Experimental results show that the proposed model can not only accurately analyze the spatiotemporal saliency based on the intensity, the texture, and the motion features, but also able to estimate the blocking artifact of distortions in comparing with Walther's and You’s models.
- 2
- 2
- 2
- Dilated kernel prediction network for single-image denoising1
A kernel prediction network that examines each pixel region and predicts unique pixel-wise kernels and constructs a compressed feature map as a substitution such that more relevant local features are collected and achieves favorable results compared with state-of-the-art methods.
- Space-time super-resolution with motion-perceptive deformable alignment1
A motion-perceptive deformable alignment network that introduces a pre-computed optical flow to improve the motion perception of the deformable aligned process and the superiority of the proposed network among both the one- and two-stage STSR methods is demonstrated.
- Multiframe interpolation for video using phase features1
A phase-feature-aided multiframe interpolation network that aims to estimate multiple in-between frames in one pass and handle challenging scenarios such as extreme light changes and occlusion and achieves marked visual improvement in challenging scenarios is presented.
- Blind watermarking algorithm in H.264 compressed domain1
To solve the problem of H.264 video copyright protection, a new blind watermarking algorithm was proposed that can effectively resist the re-coded attacks of different Quantization Parameter(QP).
- 1
- –
- Dynamic PDGAN: discriminator-boosted knowledge distillation for StyleGANs–
This work proposes to exploit the teacher discriminator to additionally supervise the student generator of StyleGANs so as to leverage the knowledge in the teacher discriminator, and can outperform original StyleGANs on several large-scale datasets.
- Contrastive visual feature filtering for generalized zero-shot learning–
This paper proposes a contrastive visual feature filtering framework (CVFF) for the generalized zero-shot learning task, eliminating redundant parts from both the real and the synthetic visual features.
- Multi-Objective Trajectory Planning with Dual-Encoder–
This paper introduces a dual-encoder based transformer model to establish a good preliminary trajectory and refined through sequential quadratic programming to improve its optimality and robustness, and shrinks the optimality gap.
- –
- Gradient Agreement Hinders the Memorization of Noisy Labels–
This work uses two synchronously trained networks to reveal that noisy labels may result in more divergent gradients when updating the parameters of two identical DNNs, and proposes a novel co-training framework named gradient agreement learning (GAL).
- FGCM: Noisy Label Learning via Fine-Grained Confidence Modeling–
A novel fine-grained confidence modeling (FGCM) framework is proposed to cluster samples into these five categories; with each cluster, FGCM decides whether to accept the cluster data as they are, accept them with label correction, or accept them as unlabeled data.
- Just noticeable disparity error-based depth coding for three-dimensional video–
A just noticeable disparity error (JNDE) measurement to describe the maximum tolerated error of depth maps is proposed and can be incorporated in any standardized video coding algorithm based on prediction and transform.
- Perceptual Video Streaming Protection Model by Priority Classification in Compress Domain–
Results of the experiment show that the proposed method outperforms former ARQ method in the objective and subjective quality of video transmission.
- Reference Viewpoints Selection for Multi-View Video Plus Depth Coding Based on the Network Bandwidth Constraint–
The results show that depending on the available network bandwidth, the MVD coding requires different selections of reference viewpoints to maximize the entire video quality of all display views.
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.