A5100417589- Advanced Vision and Imaging
- Advanced Image Processing Techniques
- Video Surveillance and Tracking Methods
- Advanced Neural Network Applications
- Image Processing Techniques and Applications
- Face recognition and analysis
- Image Enhancement Techniques
- Advanced Image and Video Retrieval Techniques
- Advanced Optical Sensing Technologies
- Computer Graphics and Visualization Techniques
- Infrared Target Detection Methodologies
- Visual Attention and Saliency Detection
- Sentiment Analysis and Opinion Mining
- Image Processing and 3D Reconstruction
- Face and Expression Recognition
- Advanced Text Analysis Techniques
- Random lasers and scattering media
- Plasma Applications and Diagnostics
- Enzyme Production and Characterization
- Video Coding and Compression Technologies
- Advanced Image Fusion Techniques
- Optical Imaging and Spectroscopy Techniques
- 3D Shape Modeling and Analysis
- Photoacoustic and Ultrasonic Imaging
- Probiotics and Fermented Foods
Southwest University
2022
Dalian University of Technology
2021
Bristol Robotics Laboratory
2016-2020
ZTE (United States)
2020
BOE Technology Group (China)
2020
Peking University Shenzhen Hospital
2019
Peking University
2018
Beihang University
2017
University of the West of England
2016
Tsinghua University
2012-2016
Super-Resolution (SR) is a fundamental computer vision task that aims to obtain high-resolution clean image from the given low-resolution counterpart. This paper reviews NTIRE 2021 Challenge on Video Super-Resolution. We present evaluation results two competition tracks as well proposed solutions. Track 1 develop conventional video SR methods focusing restoration quality. 2 assumes more challenging environment with lower frame rates, casting spatio-temporal problem. In each competition, 247...
The integration of deep generative networks into generating Computer-Aided Design (CAD) models has garnered increasing attention over recent years. Traditional methods often rely on discrete sequences parametric line/curve segments to represent sketches. Differently, we introduce RECAD, a novel framework that generates Raster sketches and 3D Extrusions for CAD models. Representing as raster images offers several advantages sequences: 1) it breaks the limitations types numbers lines/curves,...
Saliency detection aims to detect the most attractive objects in images and is widely used as a foundation for various applications. In this article, we propose novel salient object algorithm RGB-D using center-dark channel priors. First, generate an initial saliency map based on color depth of given image. Then, center dark Finally, fuse with final map. Extensive evaluations over four benchmark datasets demonstrate that our proposed method performs favorably against state-of-the-art...
This paper seeks to compare encoded features from both two-dimensional (2D) and three-dimensional (3D) face images in order achieve automatic gender recognition with high accuracy robustness. The Fisher vector encoding method is employed produce 2D, 3D, fused escalated discriminative power. For 3D analysis, a two-source photometric stereo (PS) introduced that enables surface reconstructions accurate details as well desirable efficiency. Moreover, 2D+3D imaging device, taking the PS its core,...
This paper reviews the video extreme super-resolution challenge associated with AIM 2020 workshop at ECCV 2020. Common scaling factors for learned (VSR) do not go beyond factor 4. Missing information can be restored well in this region, especially HR videos, where high-frequency content mostly consists of texture details. The task is to upscale videos an 16, which results more serious degradations that also affect structural integrity videos. A single pixel low-resolution (LR) domain...
Traditional video coding standards, such as HEVC and VVC, have achieved significant compression performance. To further improve the efficiency, a post-processing network is proposed to enhance compressed frames in this paper. Specifically, network, namely DIA_Net, contains multiple inception blocks, attention mechanism dense residual structure. The DIA_Net can efficiently extract information of scale fully exploit extracted feature image quality. In addition, integrated into latest test...
Illumination variance is one of the largest real-world problems when deploying face recognition systems. Over last few years much work has gone into development novel 3D methods to overcome this issue. Photometric stereo a well-established reconstruction technique capable recovering normals and albedo surface. Although it provides way obtain data, amount training data available captured using photometric often does not provide sufficient modelling capacity for state-of-the-art feature...
Saliency detection aims to detect the most attractive objects in images and is widely used as a foundation for various applications. In this paper, we propose novel salient object algorithm RGB-D using center-dark channel priors. First, generate an initial saliency map based on color depth of given image. Then, center dark Finally, fuse with final map. Extensive evaluations over four benchmark datasets demonstrate that our proposed method performs favorably against state-of-the-art...
Recent advances in deep generative models have led to immense progress 3D shape synthesis. While existing are able synthesize shapes represented as voxels, point-clouds, or implicit functions, these methods only indirectly enforce the plausibility of final surface. Here we present a synthesis framework (SurfGen) that directly applies adversarial training object Our approach uses differentiable spherical projection layer capture and represent explicit zero isosurface an generator functions...
Visible-infrared person re-identification (VI-ReID) aims to search the same pedestrian of interest across visible and infrared modalities. Existing models mainly focus on compensating for modality-specific information reduce modality variation. However, these methods often lead a higher computational overhead may introduce interfering when generating corresponding images or features. To address this issue, it is critical leverage pedestrian-attentive features learn modality-complete...