Xingyu Miao

Xingyu Miao

Auto-generated profile, based on publicly available information

Xingyu Miao

Durham University, Durham, United Kingdom

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

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

Representative Publications

View All Publications
15 August 2023 in IEEE Transactions on Circuits and Systems for Video Technology
Verified
Journal Article
Closed access
XMXingyu Miao
Xingyu Miao
YBYang Bai
Yang Bai
HDHaoran Duan
Haoran Duan
YHYawen Huang
Yawen Huang
FWFan Wan
Fan Wan
YZYefeng Zheng
Yefeng Zheng
Self-supervised monocular depth estimation methods typically rely on the reprojection error to capture geometric relationships between successive frames in static environments. However, this assumption does not hold in dynamic objects in scenarios, leading to errors during the view synthesis stage, such as feature mismatch and occlusion, which can significantly reduce the accuracy of the generated depth maps. To address this problem, we propose a novel dynamic cost volume that exploits residual optical flow to describe moving objects, improving incorrectly occluded regions in static cost volumes used in previous work. Nevertheless, the dynamic cost volume inevitably generates extra occlusions and noise, thus we alleviate this by designing a fusion module that makes static and dynamic cost volumes compensate for each other. In other words, occlusion from the static volume is refined by the dynamic volume, and incorrect information from the dynamic volume is eliminated by the static volume. Furthermore, we propose a pyramid distillation loss to reduce photometric error inaccuracy at low resolutions and an adaptive photometric error loss to alleviate the flow direction of the large gradient in the occlusion regions. We conducted extensive experiments on the KITTI and Cityscapes datasets, and the results demonstrate that our model outperforms previously published baselines for self-supervised monocular depth estimation.

1 July 2024 in Pattern Recognition
Verified
Journal Article
Closed access
XMXingyu Miao
Xingyu Miao
YBYang Bai
Yang Bai
HDHaoran Duan
Haoran Duan
FWFan Wan
Fan Wan
YHYawen Huang
Yawen Huang
YZYefeng Zheng
Yefeng Zheng

29 January 2025 in IEEE Transactions on Pattern Analysis and Machine Intelligence
Verified
Journal Article
Closed access
XMXingyu Miao
Xingyu Miao
HDHaoran Duan
Haoran Duan
YBYang Bai
Yang Bai
TSTejal Shah
Tejal Shah
JSJun Song
Jun Song
LSLing Shao
Ling Shao
In this work, we propose a method that leverages CLIP feature distillation, achieving efficient 3D segmentation through language guidance. Unlike previous methods that rely on multi-scale CLIP features and are limited by processing speed and storage requirements, our approach aims to streamline the workflow by directly and effectively distilling dense CLIP features, thereby achieving precise segmentation of 3D scenes using text. To achieve this, we introduce an adapter module and mitigate the noise issue in the dense CLIP feature distillation process through a self-cross-training strategy. Moreover, to enhance the accuracy of segmentation edges, this work presents a low-rank transient query attention mechanism. To ensure the consistency of segmentation for similar colors under different viewpoints, we convert the segmentation task into a classification task through label volume, which significantly improves the consistency of segmentation in color-similar areas. We also propose a simplified text augmentation strategy to alleviate the issue of ambiguity in the correspondence between CLIP features and text. Extensive experimental results show that our method surpasses current state-of-the-art technologies in both training speed and performance. Our code is available on: https://github.com/xingy038/Laser.git.

21 January 2026 in Pattern Recognition
Verified
Journal Article
Open access
XMXingyu Miao
Xingyu Miao
YBYang Bai
Yang Bai
HDHaoran Duan
Haoran Duan
FWFan Wan
Fan Wan
YHYawen Huang
Yawen Huang
YZYefeng Zheng
Yefeng Zheng

30 October 2020 in Journal of Electrical and Computer Engineering
Verified
Journal Article
Open access
XMXingyu Miao
Xingyu Miao
YGYongqi Ge
Yongqi Ge
Energy management for energy harvesting-based embedded systems (EHES) is an emerging field, which aims to collect renewable energy from the environment to power an embedded system. In this work, we use the systematic mapping method to study the relevant literature, with the objective of exploring and analysing the state of the art in energy management for EHES, as well as to provide assistance for subsequent literature reviews. To this end, we conducted extensive searches to find articles related to energy harvesting, embedded systems, energy consumption, and energy management. We searched for papers from January 2005 to July 2019 from three mainstream databases, ACM, IEEE Xplore, and Web of Science, and found more than 3000 papers about EHES. Finally, we selected 142 eligible papers. We have completed the system mapping research from five aspects, namely, (1) research type (validation research, evaluation research, solution proposal, philosophical paper, opinion, and experience), (2) research goals (application or theory), (3) application scenarios, (4) tools or methods, and (5) paper distribution, such as publication year and authors’ nationality. The results showed that the major research type of the EHES papers is validation research, accounting for 65%, which indicated research is still in the theoretical stage and many researchers focus on how to improve the efficiency of harvesting energy, develop a reasonable energy supply plan, and adapt EHES for real-world requirements. Furthermore, this work reviews the tools used for EHES. As the future development direction, it is indispensable to provide tools to EHES for research, testing, development, and so on. The results of our analysis provide significant contributions to understanding the existing knowledge and highlighting potential future research opportunities in the EHES field.

1 January 2026 in IEEE Transactions on Circuits and Systems for Video Technology
Verified
Journal Article
Closed access
XMXingyu Miao
Xingyu Miao
HDHaoran Duan
Haoran Duan
ZOZizhou Ouyang
Zizhou Ouyang
RHRunze Hu
Runze Hu
TSTejal Shah
Tejal Shah
RRRajiv Ranjan
Rajiv Ranjan