Xinyao Wang

Xinyao Wang

Auto-generated profile, based on publicly available information

Xinyao Wang

College of Automation Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing, China

Research Focus & Metrics

Academic Career Length

8Years (2019-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
19 November 2020
Verified
Book Chapter
Closed access
PDF file
XWXinyao Wang
Xinyao Wang
TYTaiping Yao
Taiping Yao
SDShouhong Ding
Shouhong Ding
LMLizhuang Ma
Lizhuang Ma

19 August 2025 in IEEE Internet of Things Journal
Verified
Journal Article
Closed access
YJYuqian Jing
Yuqian Jing
Yuben QuYuben Qu
Yuben Qu
ZQZhen Qin
Zhen Qin
CDChao Dong
Chao Dong
FZFuhui Zhou
Fuhui Zhou
QWQihui Wu
Qihui Wu
Federated learning (FL), as an innovative paradigm of distributed learning, provides reliable support for the growing edge intelligence (EI). The limitations of traditional FL’s reliance on ground base stations (BSs) make the development of aerial server autonomous aerial vehicles (AAVs) inevitable, thereby developing the air–ground integrated FL (AGIFL). However, current efforts predominately focus on static offline training based on existing datasets and some efforts consider online training in dynamic sample environments, where new samples need to be fully pretrained to determine sample quality. To this end, we study how to realize high-performance of FL in dynamic environment without training all samples. Specifically, we formulate a joint optimization problem for sample selection, AAV deployment, and resource distribution aiming to minimize the tradeoff between the user energy consumption and FL performance. To address the optimization problem without explicit expression, we employ meta-learning to derive an upper bound on the gradient norm of the loss function to evaluate learning performance, and describe how time-varying small-batch ratios affect this bound. Then, we propose an optimization algorithm that ensures convergence, capitalizing on the block coordinate descent techniques. To demonstrate the efficacy of our algorithm, we conduct both extensive simulations and proof-of-concept field experiments. The findings indicate an average improvement of approximately 39% in reducing the objective value when compared to the benchmarks.

1 December 2019
Verified
Conference Paper
Closed access
XWXinyao Wang
Xinyao Wang
JWJing Wang
Jing Wang
ZYZhaoyu Yan
Zhaoyu Yan
MLMin Liu
Min Liu
KQKai Qian
Kai Qian
Head-related transfer function(HRTF) represents the transfer function from audio source in the spatial field to ear drums of human, which is commonly used for rendering virtual sound sources. The basic measurement method to get high-resolution HRTF is limited for its time-consuming and complicated procedure. Therefore, some reconstructed methods based on low-resolution HRTFs are presented, and the most classical two methods are Surface Spherical Harmonics(SSH) and Linear Interpolation(LI). The main work of this paper is to analyze the rendering performance of HRTF under these two methods. Objective and subjective experiments are conducted on MIT database. Through comparing the Mean Square Error(MSE) and Spectral Distortion(SD) between reconstructed and original spectrum, the objective evaluation shows that the effect of SSH on the shadowed side and backward is not as stable as linear interpolation. Furthermore, subjective evaluation indicates that SSH has a better performance in localization perception with low quality degradation. Meanwhile, the performance related to accuracy, feasibility, expansibility is analyzed comprehensively.

1 July 2021 in arXiv
Verified
Preprint
Open access
PDF file
DHDexiang Hong
Dexiang Hong
CLCongcong Li
Congcong Li
LWLongyin Wen
Longyin Wen
XWXinyao Wang
Xinyao Wang
LZLibo Zhang
Libo Zhang
This report presents the approach used in the submission of Generic Event Boundary Detection (GEBD) Challenge at CVPR21. In this work, we design a Cascaded Temporal Attention Network (CASTANET) for GEBD, which is formed by three parts, the backbone network, the temporal attention module, and the classification module. Specifically, the Channel-Separated Convolutional Network (CSN) is used as the backbone network to extract features, and the temporal attention module is designed to enforce the network to focus on the discriminative features. After that, the cascaded architecture is used in the classification module to generate more accurate boundaries. In addition, the ensemble strategy is used to further improve the performance of the proposed method. The proposed method achieves 83.30% F1 score on Kinetics-GEBD test set, which improves 20.5% F1 score compared to the baseline method. Code is available at https://github.com/DexiangHong/Cascade-PC.

29 March 2022 in arXiv
Verified
Preprint
Open access
PDF file
CLCongcong Li
Congcong Li
XWXinyao Wang
Xinyao Wang
LWLongyin Wen
Longyin Wen
DHDexiang Hong
Dexiang Hong
TLTiejian Luo
Tiejian Luo
LZLibo Zhang
Libo Zhang
Generic event boundary detection aims to localize the generic, taxonomy-free event boundaries that segment videos into chunks. Existing methods typically require video frames to be decoded before feeding into the network, which demands considerable computational power and storage space. To that end, we propose a new end-to-end compressed video representation learning for event boundary detection that leverages the rich information in the compressed domain, i.e., RGB, motion vectors, residuals, and the internal group of pictures (GOP) structure, without fully decoding the video. Specifically, we first use the ConvNets to extract features of the I-frames in the GOPs. After that, a light-weight spatial-channel compressed encoder is designed to compute the feature representations of the P-frames based on the motion vectors, residuals and representations of their dependent I-frames. A temporal contrastive module is proposed to determine the event boundaries of video sequences. To remedy the ambiguities of annotations and speed up the training process, we use the Gaussian kernel to preprocess the ground-truth event boundaries. Extensive experiments conducted on the Kinetics-GEBD dataset demonstrate that the proposed method achieves comparable results to the state-of-the-art methods with $4.5\times$ faster running speed.