Search

Your search keyword '"Cui Qingzhe"' showing total 20 results
Start Over Author "Cui Qingzhe"
20 results on '"Cui Qingzhe"'

1. Group Multi-View Transformer for 3D Shape Analysis with Spatial Encoding

Author

Xu, Lixiang, Cui, Qingzhe, Hong, Richang, Xu, Wei, Chen, Enhong, Yuan, Xin, Li, Chenglong, and Tang, Yuanyan

Subjects
Computer Science - Computer Vision and Pattern Recognition, Computer Science - Artificial Intelligence, 68, I.2.10
Abstract

In recent years, the results of view-based 3D shape recognition methods have saturated, and models with excellent performance cannot be deployed on memory-limited devices due to their huge size of parameters. To address this problem, we introduce a compression method based on knowledge distillation for this field, which largely reduces the number of parameters while preserving model performance as much as possible. Specifically, to enhance the capabilities of smaller models, we design a high-performing large model called Group Multi-view Vision Transformer (GMViT). In GMViT, the view-level ViT first establishes relationships between view-level features. Additionally, to capture deeper features, we employ the grouping module to enhance view-level features into group-level features. Finally, the group-level ViT aggregates group-level features into complete, well-formed 3D shape descriptors. Notably, in both ViTs, we introduce spatial encoding of camera coordinates as innovative position embeddings. Furthermore, we propose two compressed versions based on GMViT, namely GMViT-simple and GMViT-mini. To enhance the training effectiveness of the small models, we introduce a knowledge distillation method throughout the GMViT process, where the key outputs of each GMViT component serve as distillation targets. Extensive experiments demonstrate the efficacy of the proposed method. The large model GMViT achieves excellent 3D classification and retrieval results on the benchmark datasets ModelNet, ShapeNetCore55, and MCB. The smaller models, GMViT-simple and GMViT-mini, reduce the parameter size by 8 and 17.6 times, respectively, and improve shape recognition speed by 1.5 times on average, while preserving at least 90% of the classification and retrieval performance. The code is available at https://github.com/bigdata-graph/GMViT., Comment: 13pages, 8 figuers

Published
2023

13. Long short-term memory network-based wastewater quality prediction model with sparrow search algorithm.

Author

Li, Guobin, Cui, Qingzhe, Wei, Shengnan, Wang, Xiaofeng, Xu, Lixiang, He, Lixin, Kwong, Timothy C. H., and Tang, Yuanyan

Subjects
*SEARCH algorithms, *PREDICTION models, *OPTIMIZATION algorithms, *PRINCIPAL components analysis, *TIME complexity, *WATER quality monitoring
Abstract

The wastewater treatment process is characterized by uncertainty, non-linearity, time delay and complexity, and is susceptible to many dynamic factors. Since some key water quality parameters are not available in real time, a Long Short-Term Memory (LSTM) network water quality prediction model based on sparrow search algorithm (SSA-LSTM) and attention mechanism is proposed to solve the problem. In this model, we take historical data as input, constructs models to learn the feature of internal dynamic changes, introduces the attention mechanism, assigns different weights to the hidden state of the LSTM network by mapping weightings with the learning parameter matrix, and uses the SSA to select the optimal hyperparameters for prediction. As high-latitude feature vectors are subject to the curse of dimension, a PCA-LSTM model is further proposed to apply the Principal Component Analysis (PCA) method to the SSA-LSTM model to reduce the dimensionality of the original data. The SSA-LSTM model without the PCA method (NPCA-LSTM) and the PCA-LSTM model are applied to predict wastewater quality and the PCA-LSTM model shows higher predictive ability. [ABSTRACT FROM AUTHOR]

Published
2023
Full Text
View/download PDF

14. Effects of Dentin Ablation by a Q-Switching Er:YSGG Laser with a High Pulse Repetition Rate

Author

Hu Shuwu, Cui Qingzhe, Wu Xianyou, Xiong Zhengdong, Jiang Haihe, Jiang Jiantao, Cheng Tingqing, Wei Meng'en, and Wang Li

Subjects
Pulse repetition frequency, Materials science, Repetition (rhetorical device), Pulse (signal processing), business.industry, medicine.medical_treatment, Biomedical Engineering, Lasers, Solid-State, Laser, Ablation, Q-switching, law.invention, stomatognathic diseases, medicine.anatomical_structure, Optics, stomatognathic system, law, Dentin, medicine, Radiology, Nuclear Medicine and imaging, business
Abstract

Objective: The aim of the study was to evaluate the characteristics of dentin ablation with a high pulse repetition rate Q-switching 2.79 μm laser. Materials and methods: Dentin was ablated using a...

Published
2021

15. 100–300 Hz repetition-rate acousto-optic Q-switched 2.79 μm Er:YSGG laser side-pumped by laser-diode

Author

Hu Shuwu, Li Wang, Xiong Zhengdong, Wu Xianyou, Wei Meng'en, Cui Qingzhe, Jiang Haihe, Jiang Jiantao, and Cheng Tingqing

Subjects
Materials science, Repetition (rhetorical device), Laser diode, business.industry, Pulse duration, 02 engineering and technology, 021001 nanoscience & nanotechnology, Condensed Matter Physics, Laser, 01 natural sciences, Atomic and Molecular Physics, and Optics, Electronic, Optical and Magnetic Materials, Power (physics), law.invention, Pulse (physics), 010309 optics, Crystal, Optics, law, Optical cavity, 0103 physical sciences, 0210 nano-technology, business
Abstract

An LD-pumped high-repetition-rate Er:YSGG laser utilizing a TeO2 crystal as an acousto-optic Q-switch is proposed and demonstrated. The thermal lensing effect caused by the high-repetition-rate pump is compensated by a plano-convex laser resonator. As a result, a Q-switched 2.79 μm laser with a repetition rate of 100–300 Hz is obtained. When a 100 Hz repetition rate, a 1 mJ pulse energy and a 76 ns pulse duration are employed, the corresponding peak power is 13.2 kW. To the best of our knowledge, this laser offers the highest pulse energy among Q-switched 2.79 μm lasers with pulse repetition rates above one hundred hertz.

Published
2019

19. Effect of Water Mist on Threshold and Damage of Dentin Ablation with Q-Switched Er:YAG Laser

Author

Ling Lin, Cui Qingzhe, Wu Xianyou, W Ei Meng'En, Cheng Tingqing, and Jiang Haihe

Subjects
Thesaurus (information retrieval), Materials science, business.industry, medicine.medical_treatment, Mist, Ablation, Atomic and Molecular Physics, and Optics, Electronic, Optical and Magnetic Materials, medicine.anatomical_structure, Dentin, medicine, Optoelectronics, Electrical and Electronic Engineering, business, Er:YAG laser
Published
2016

20. Generation of sub-100 fs vortices from a Kerr-lens mode-locked Hermite-Gaussian Yb:CALGO oscillator.

Author

Wu F, Liu H, Yan L, Liu H, Cui Q, Dai K, Li H, Qin Y, Wang Q, and Zhang J

Abstract

Mode-locked oscillators with high-order transverse modes are excellent platforms for generating femtosecond optical vortices with high average power and good propagation stability. These have important applications in diverse fields such as optical communication, strong-field physics, and laser processing. So far, generating vortex pulses with ultrashort pulse duration remains a challenge. In this Letter, we report a Kerr-lens mode-locked Yb:CALGO laser oscillator delivering Hermite-Gaussian (HG) pulses with a pulse duration of 86 fs using a non-collinear pumping technique. 91 fs optical vortex pulses were generated by using a cylindrical-lens mode converter. To the best of our knowledge, this is the shortest pulse duration ever obtained from a diode pumped solid-state mode-locked oscillator with a pure high-order Hermite-Gaussian mode. The phase structures of the generated femtosecond vortices are characterized.

Published
2024
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results