no code implementations • 1 Feb 2023 • Yong Xiao, Rong Xia, Yingyu Li, Guangming Shi, Diep N. Nguyen, Dinh Thai Hoang, Dusit Niyato, Marwan Krunz
FS-GAN is composed of multiple distributed Generative Adversarial Networks (GANs), with a set of generators, each being designed to generate synthesized data samples following the distribution of an individual service traffic, and each discriminator being trained to differentiate the synthesized data samples and the real data samples of a local dataset.
no code implementations • 27 Jan 2023 • Zhimin Lu, Yong Xiao, Zijian Sun, Yingyu Li, Guangming Shi, Xianfu Chen, Mehdi Bennis, H. Vincent Poor
In this paper, we consider the implicit semantic communication problem in which hidden relations and closely related semantic terms that cannot be recognized from the source signals need to also be delivered to the destination user.
no code implementations • 26 Jan 2023 • Yong Xiao, Xiaohan Zhang, Guangming Shi, Marwan Krunz, Diep N. Nguyen, Dinh Thai Hoang
A joint optimization algorithm is proposed to minimize the overall time consumption of model training by selecting participating edge servers, local epoch number.
no code implementations • 13 Nov 2022 • Yubo Dong, Dahua Gao, Tian Qiu, Yuyan Li, Minxi Yang, Guangming Shi
However, in the data subproblem, the used sensing matrix is ill-suited for the real degradation process due to the device errors caused by phase aberration, distortion; in the prior subproblem, it is important to design a suitable model to jointly exploit both spatial and spectral priors.
1 code implementation • 28 Oct 2022 • Yong Xiao, Zijian Sun, Guangming Shi, Dusit Niyato
A federated GCN-based collaborative reasoning solution is proposed to allow multiple edge servers to jointly construct a shared semantic interpretation model based on decentralized knowledge datasets.
1 code implementation • 12 Apr 2022 • Yang Li, Ji Chen, Fu Li, Boxun Fu, Hao Wu, Youshuo Ji, Yijin Zhou, Yi Niu, Guangming Shi, Wenming Zheng
GMSS has the ability to learn more general representations by integrating multiple self-supervised tasks, including spatial and frequency jigsaw puzzle tasks, and contrastive learning tasks.
no code implementations • 9 Mar 2022 • Jingxuan Chai, Guangming Shi
Specifically, under our framework, we introduce a more generic embedding method, ModulE, which projects entities to a module.
no code implementations • 15 Dec 2021 • Yufan Zhu, Weisheng Dong, Leida Li, Jinjian Wu, Xin Li, Guangming Shi
In this work, we introduce uncertainty-driven loss functions to improve the robustness of depth completion and handle the uncertainty in depth completion.
no code implementations • 14 Dec 2021 • Yijin Zhou, Fu Li, Yang Li, Youshuo Ji, Guangming Shi, Wenming Zheng, Lijian Zhang, Yuanfang Chen, Rui Cheng
Moreover, motivated by the observation of the relationship between coarse- and fine-grained emotions, we adopt a dual-head module that enables the PGCN to progressively learn more discriminative EEG features, from coarse-grained (easy) to fine-grained categories (difficult), referring to the hierarchical characteristic of emotion.
no code implementations • NeurIPS 2021 • Qian Ning, Weisheng Dong, Xin Li, Jinjian Wu, Guangming Shi
Specifically, we introduce variance estimation characterizing the uncertainty on a pixel-by-pixel basis into SISR solutions so the targeted pixels in a high-resolution image (mean) and their corresponding uncertainty (variance) can be learned simultaneously.
no code implementations • 21 Apr 2021 • Chenxin Xu, Rong Xia, Yong Xiao, Yingyu Li, Guangming Shi, Kwang-cheng Chen
With the fast growing demand on new services and applications as well as the increasing awareness of data protection, traditional centralized traffic classification approaches are facing unprecedented challenges.
1 code implementation • 12 Apr 2021 • Pengfei Wang, Chengquan Zhang, Fei Qi, Shanshan Liu, Xiaoqiang Zhang, Pengyuan Lyu, Junyu Han, Jingtuo Liu, Errui Ding, Guangming Shi
With a PG-CTC decoder, we gather high-level character classification vectors from two-dimensional space and decode them into text symbols without NMS and RoI operations involved, which guarantees high efficiency.
Ranked #1 on
Scene Text Detection
on ICDAR 2015
(Accuracy metric)
no code implementations • 6 Apr 2021 • Qian Ning, Weisheng Dong, Xin Li, Jinjian Wu, Leida Li, Guangming Shi
Similar to the success of NAS in high-level vision tasks, it is possible to find a memory and computationally efficient solution via NAS with highly competent denoising performance.
1 code implementation • CVPR 2021 • Tao Huang, Weisheng Dong, Xin Yuan, Jinjian Wu, Guangming Shi
Different from existing GSM models using hand-crafted scale priors (e. g., the Jeffrey's prior), we propose to learn the scale prior through a deep convolutional neural network (DCNN).
no code implementations • 29 Jan 2021 • Guangming Shi, Dahua Gao, Xiaodan Song, Jingxuan Chai, Minxi Yang, Xuemei Xie, Leida Li, Xuyang Li
In this article, we deploy semantics to solve the spectrum and power bottleneck and propose a first understanding and then transmission framework with high semantic fidelity.
Networking and Internet Architecture
no code implementations • 16 Jan 2021 • Yang Jiao, Guangming Shi, Trac D. Tran
In this paper, we discover that the lost information is related to a large quantity of motion features (more than 40%) computed from the popular discriminative cost-volume feature would completely vanish due to invalid sampling, leading to the low efficiency of optical flow learning.
1 code implementation • ICCV 2021 • Pengfei Chen, Leida Li, Jinjian Wu, Weisheng Dong, Guangming Shi
From this adaptation, we split the data in target domain into confident and uncertain subdomains using the proposed uncertainty-based ranking function, through measuring their prediction confidences.
no code implementations • 16 Dec 2020 • Jianan Li, Xuemei Xie, Zhifu Zhao, Yuhan Cao, Qingzhe Pan, Guangming Shi
Specifically, the constructed temporal relation graph explicitly builds connections between semantically related temporal features to model temporal relations between both adjacent and non-adjacent time steps.
no code implementations • 25 Nov 2020 • Yong Xiao, Yingyu Li, Guangming Shi, H. Vincent Poor
The data uploading performance of IoT network and the computational capacity of edge servers are entangled with each other in influencing the FL model training process.
no code implementations • CVPR 2021 • Yang Jiao, Trac D. Tran, Guangming Shi
This paper addresses the challenging unsupervised scene flow estimation problem by jointly learning four low-level vision sub-tasks: optical flow $\textbf{F}$, stereo-depth $\textbf{D}$, camera pose $\textbf{P}$ and motion segmentation $\textbf{S}$.
no code implementations • 16 Nov 2020 • Yang Jiao, Yi Niu, Trac D. Tran, Guangming Shi
In 2D+3D facial expression recognition (FER), existing methods generate multi-view geometry maps to enhance the depth feature representation.
no code implementations • 1 Oct 2020 • Yong Xiao, Guangming Shi, Yingyu Li, Walid Saad, H. Vincent Poor
Edge intelligence, also called edge-native artificial intelligence (AI), is an emerging technological framework focusing on seamless integration of AI, communication networks, and mobile edge computing.
no code implementations • 21 Sep 2020 • Yang Li, Boxun Fu, Fu Li, Guangming Shi, Wenming Zheng
So it is necessary to give more attention to the EEG samples with strong transferability rather than forcefully training a classification model by all the samples.
no code implementations • 14 Sep 2020 • Qian Ning, Weisheng Dong, Guangming Shi, Leida Li, Xin Li
Deep neural networks (DNNs) based methods have achieved great success in single image super-resolution (SISR).
no code implementations • 26 Apr 2020 • Yong Xiao, Guangming Shi, Marwan Krunz
One of the key challenges is the difficulty to implement distributed AI across a massive number of heterogeneous devices.
1 code implementation • CVPR 2020 • Hancheng Zhu, Leida Li, Jinjian Wu, Weisheng Dong, Guangming Shi
The underlying idea is to learn the meta-knowledge shared by human when evaluating the quality of images with various distortions, which can then be adapted to unknown distortions easily.
1 code implementation • 15 Aug 2019 • Pengfei Wang, Chengquan Zhang, Fei Qi, Zuming Huang, Mengyi En, Junyu Han, Jingtuo Liu, Errui Ding, Guangming Shi
Detecting scene text of arbitrary shapes has been a challenging task over the past years.
Ranked #18 on
Scene Text Detection
on ICDAR 2015
no code implementations • 29 Sep 2018 • Guangming Shi, Zhongqiang Zhang, Dahua Gao, Xuemei Xie, Yihao Feng, Xinrui Ma, Danhua Liu
Besides, to enhance the recognition ability of the semantic tree in aspects of the diversity, randomicity and variability, we use the traditional neural network to aid the semantic tree to learn some indescribable features.
no code implementations • 18 Jul 2018 • Fangfang Wu, Weisheng Dong, Guangming Shi, Xin Li
State-of-the-art approaches toward image restoration can be classified into model-based and learning-based.
no code implementations • 13 Feb 2018 • Weishong Dong, Ming Yuan, Xin Li, Guangming Shi
Image demosaicing - one of the most important early stages in digital camera pipelines - addressed the problem of reconstructing a full-resolution image from so-called color-filter-arrays.
1 code implementation • 1 Feb 2018 • Jiang Du, Xuemei Xie, Chenye Wang, Guangming Shi
In detail, we employ perceptual loss, defined on feature level, to enhance the structure information of the recovered images.
1 code implementation • 1 Feb 2018 • Xuemei Xie, Chenye Wang, Jiang Du, Guangming Shi
In measurement part, the input image is adaptively measured block by block to acquire a group of measurements.
no code implementations • 31 Jan 2018 • Xiaotong Lu, Weisheng Dong, Peiyao Wang, Guangming Shi, Xuemei Xie
Instead of reconstructing individual blocks, the whole image is reconstructed from the linear convolutional measurements.
no code implementations • 21 Jan 2018 • Weisheng Dong, Peiyao Wang, Wotao Yin, Guangming Shi, Fangfang Wu, Xiaotong Lu
Then, the iterative process is unfolded into a deep neural network, which is composed of multiple denoisers modules interleaved with back-projection (BP) modules that ensure the observation consistencies.
1 code implementation • 21 Nov 2017 • Jiang Du, Xuemei Xie, Chenye Wang, Guangming Shi, Xun Xu, Yu-Xiang Wang
Recently, deep learning methods have made a significant improvement in compressive sensing image reconstruction task.
no code implementations • 5 Oct 2017 • Xuemei Xie, Chenye Wang, Shu Chen, Guangming Shi, Zhifu Zhao
Experiments show that the system can achieve a 99% accuracy and real-time (25FPS) detection with strong robustness in complex environments.
1 code implementation • 23 Sep 2017 • Xuemei Xie, Yu-Xiang Wang, Guangming Shi, Chenye Wang, Jiang Du, Zhifu Zhao
In this paper, we propose an adaptive measurement network in which measurement is obtained by learning.
1 code implementation • 15 Sep 2017 • Guimei Cao, Xuemei Xie, Wenzhe Yang, Quan Liao, Guangming Shi, Jinjian Wu
We propose a multi-level feature fusion method for introducing contextual information in SSD, in order to improve the accuracy for small objects.
no code implementations • NeurIPS 2016 • Yongbo Li, Weisheng Dong, Xuemei Xie, Guangming Shi, Xin Li, Donglai Xu
More specifically, the parametric sparse prior of the desirable high-resolution (HR) image patches are learned from both the input low-resolution (LR) image and a training image dataset.
no code implementations • ICCV 2015 • Weisheng Dong, Guangyu Li, Guangming Shi, Xin Li, Yi Ma
Patch-based low-rank models have shown effective in exploiting spatial redundancy of natural images especially for the application of image denoising.
no code implementations • ICCV 2015 • Yongbo Li, Weisheng Dong, Guangming Shi, Xuemei Xie
Existing approaches toward Image super-resolution (SR) is often either data-driven (e. g., based on internet-scale matching and web image retrieval) or model-based (e. g., formulated as an Maximizing a Posterior estimation problem).
no code implementations • CVPR 2015 • Lizhi Wang, Zhiwei Xiong, Dahua Gao, Guangming Shi, Wen-Jun Zeng, Feng Wu
We propose a novel dual-camera design to acquire 4D high-speed hyperspectral (HSHS) videos with high spatial and spectral resolution.