BZPEERPapers, people, evidence and research paths
Menu

Discover

SearchFind papers, people, topics, institutions and opportunities from one place.GraphSee how papers, authors, topics and evidence connect.

Research tools

HomeContinue from the papers, people and topics that matter now.CollectionsTurn saved papers into reading lists, evidence maps and shareable context.ResearchReconstruct the path that led you from search to paper to question.CollaborationMove from a paper, topic or expert request into a focused thread.PublishCreate full-text research articles with metadata, formulas and versions.

Network

ExpertsFind specialists whose work and reviews explain why they can help.CommunitiesJoin research groups organized around papers, topics and shared evidence.

Opportunities

ListingsBrowse roles, calls and collaborations connected to real research context.

Account

Log inReturn to your saved papers, graph views and active threads.Create accountStart saving papers, building a research trail and joining teams.
Log inCreate account

Source author record

Zhiyang Liu

Zhiyang Liu appears in the imported research catalog. Authorship, coauthor and topic links are available while profile ownership is still unclaimed.

ResearcherUnclaimed source record
Computer Visioneess.IVeess.SPInformation TheoryMachine Learningmath.IT

Catalog footprint

What is connected

6works
6topics
4close collaborators

Actions

Connect this record

Log in to claim
Open graphBrowse works

Research graph

See the researcher in context

Open full explorer

Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.

Published work

6 published item(s)

preprint2026arXiv

3D Reconstruction Techniques in the Manufacturing Domain: Applications, Research Opportunities and Use Cases

This comprehensive review examines the evolution and the current state of the art in three-dimensional (3D) reconstruction techniques in manufacturing applications. The analysis covers both traditional approaches and emerging deep learning methods, showing a critical research gap in unified 3d reconstruction frameworks. Through systematic review of 106 recent publications, we classify reconstruction techniques into three primary categories: data acquisition, point cloud generation, post-processing and applications. Non-contact methods, particularly structured light scanning and stereo vision, have shown significant adoption in manufacturing, with 47% of surveyed applications focusing on quality inspection. The integration of deep learning has enhanced reconstruction accuracy and processing speed, particularly in feature extraction and matching. Key applications span design and development (13%), machining (8%), process (17%), assembly (22%), and quality inspection (40%). While current technologies achieve sub-millimeter accuracy in controlled environments, challenges persist in handling reflective surfaces and dynamic environments. Our findings indicate a trend toward hybrid systems combining multiple sensor types and processing methods to overcome individual limitations. This survey provides a structured framework for understanding current capabilities and future directions in manufacturing-focused 3D reconstruction.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Fast and Accurate Optical Fiber Channel Modeling Using Generative Adversarial Network

In this work, a new data-driven fiber channel modeling method, generative adversarial network (GAN) is investigated to learn the distribution of fiber channel transfer function. Our investigation focuses on joint channel effects of attenuation, chromic dispersion, self-phase modulation (SPM), and amplified spontaneous emission (ASE) noise. To achieve the success of GAN for channel modeling, we modify the loss function, design the condition vector of input and address the mode collapse for the long-haul transmission. The effective architecture, parameters, and training skills of GAN are also displayed in the paper. The results show that the proposed method can learn the accurate transfer function of the fiber channel. The transmission distance of modeling can be up to 1000 km and can be extended to arbitrary distance theoretically. Moreover, GAN shows robust generalization abilities under different optical launch powers, modulation formats, and input signal distributions. Comparing the complexity of GAN with the split-step Fourier method (SSFM), the total multiplication number is only 2% of SSFM and the running time is less than 0.1 seconds for 1000-km transmission, versus 400 seconds using the SSFM under the same hardware and software conditions, which highlights the remarkable reduction in complexity of the fiber channel modeling.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

A BeiDou Signal Acquisition Approach Using Variable Length Data Accumulation based on Signal Delay and Multiplication

The secondary modulation with the NeumannHoffman code increases the possibility of bit sign transition. Unlike other GNSS signals, there is no pilot component for synchronization in BeiDou B1/B3 signals, which increases the complexity in acquisition. A previous study has shown that the delay and multiplication (DAM) method is able to eliminate the bit sign transition problem, but it only applies to pretty strong signals. In this paper, a DAM-based BeiDou signal acquisition approach, called variable length data accumulation (VLDA), is proposed to acquire weak satellite signals. Firstly, the performance of DAM method versus the different delays is analyzed. The DAM operation not only eliminates bit sign transition, but it also increases noise power. Secondly, long-term signal is periodically accumulated to improve signal intensity in order to acquire weak signals. While considering the Doppler frequency shift of ranging codes, the signal length must be compensated before accumulating long-term signal. Finally, the fast-Fourier-transform based parallel code phase algorithm are used for acquisition. The simulation results indicate that the proposed VLDA method has better acquisition sensitivity than traditional non-coherent integration method under the same calculation amount. The VLDA method only requires approximately 27.5% of calculations to achieve the same acquisition sensitivity (35 dBHz). What is more, the actual experimental results verify the feasibility of the VLDA method. It can be concluded that the proposed approach is an effective and feasible method for solving the bit sign transition problem.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Automated Segmentation of Brain Gray Matter Nuclei on Quantitative Susceptibility Mapping Using Deep Convolutional Neural Network

Abnormal iron accumulation in the brain subcortical nuclei has been reported to be correlated to various neurodegenerative diseases, which can be measured through the magnetic susceptibility from the quantitative susceptibility mapping (QSM). To quantitively measure the magnetic susceptibility, the nuclei should be accurately segmented, which is a tedious task for clinicians. In this paper, we proposed a double-branch residual-structured U-Net (DB-ResUNet) based on 3D convolutional neural network (CNN) to automatically segment such brain gray matter nuclei. To better tradeoff between segmentation accuracy and the memory efficiency, the proposed DB-ResUNet fed image patches with high resolution and the patches with low resolution but larger field of view into the local and global branches, respectively. Experimental results revealed that by jointly using QSM and T$_\text{1}$ weighted imaging (T$_\text{1}$WI) as inputs, the proposed method was able to achieve better segmentation accuracy over its single-branch counterpart, as well as the conventional atlas-based method and the classical 3D-UNet structure. The susceptibility values and the volumes were also measured, which indicated that the measurements from the proposed DB-ResUNet are able to present high correlation with values from the manually annotated regions of interest.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Automatic acute ischemic stroke lesion segmentation using semi-supervised learning

Ischemic stroke is a common disease in the elderly population, which can cause long-term disability and even death. However, the time window for treatment of ischemic stroke in its acute stage is very short. To fast localize and quantitively evaluate the acute ischemic stroke (AIS) lesions, many deep-learning-based lesion segmentation methods have been proposed in the literature, where a deep convolutional neural network (CNN) was trained on hundreds of fully labeled subjects with accurate annotations of AIS lesions. Despite that high segmentation accuracy can be achieved, the accurate labels should be annotated by experienced clinicians, and it is therefore very time-consuming to obtain a large number of fully labeled subjects. In this paper, we propose a semi-supervised method to automatically segment AIS lesions in diffusion weighted images and apparent diffusion coefficient maps. By using a large number of weakly labeled subjects and a small number of fully labeled subjects, our proposed method is able to accurately detect and segment the AIS lesions. In particular, our proposed method consists of three parts: 1) a double-path classification net (DPC-Net) trained in a weakly-supervised way is used to detect the suspicious regions of AIS lesions; 2) a pixel-level K-Means clustering algorithm is used to identify the hyperintensive regions on the DWIs; and 3) a region-growing algorithm combines the outputs of the DPC-Net and the K-Means to obtain the final precise lesion segmentation. In our experiment, we use 460 weakly labeled subjects and 15 fully labeled subjects to train and fine-tune the proposed method. By evaluating on a clinical dataset with 150 fully labeled subjects, our proposed method achieves a mean dice coefficient of 0.642, and a lesion-wise F1 score of 0.822.

0 citations0 reviewsNo code linkedOpen access
preprint2014arXiv

A Comparative Study of Downlink MIMO Cellular Networks with Co-located and Distributed Base-Station Antennas

Despite the common belief that substantial capacity gains can be achieved by using more antennas at the base-station (BS) side in cellular networks, the effect of BS antenna topology on the capacity scaling behavior is little understood. In this paper, we present a comparative study on the ergodic capacity of a downlink single-user multiple-input-multiple-output (MIMO) system where BS antennas are either co-located at the center or grouped into uniformly distributed antenna clusters in a circular cell. By assuming that the number of BS antennas and the number of user antennas go to infinity with a fixed ratio $L\gg 1$, the asymptotic analysis reveals that the average per-antenna capacities in both cases logarithmically increase with $L$, but in the orders of $\log_2 L$ and $\tfracα{2}\log_2 L$, for the co-located and distributed BS antenna layouts, respectively, where $α>2$ denotes the path-loss factor. The analysis is further extended to the multi-user case where a 1-tier (7-cell) MIMO cellular network with $K\gg 1$ uniformly distributed users in each cell is considered. By assuming that the number of BS antennas and the number of user antennas go to infinity with a fixed ratio $L\gg K$, an asymptotic analysis is presented on the downlink rate performance with block diagonalization (BD) adopted at each BS. It is shown that the average per-antenna rates with the co-located and distributed BS antenna layouts scale in the orders of $\log_2 \tfrac{L}{K}$ and $\log_2 \frac{(L-K+1)^{α/2}}{K}$, respectively. The rate performance of MIMO cellular networks with small cells is also discussed, which highlights the importance of employing a large number of distributed BS antennas for the next-generation cellular networks.

0 citations0 reviewsNo code linkedOpen access