A High-Speed, Real-Time Vision System for Texture Tracking and Thread Counting

Hu, Yuting; Long, Zhiling; AlRegib, Ghassan

doi:10.1109/lsp.2018.2825309

Cited by 11 publications

(5 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A typical feature of deep learning compared to shallow learning is that training requires large training datasets. e collection of datasets is a large workload, usually using data expansion methods [12,13]. For example, image data trimming operations, digital image enhancement operations to improve image quality, these operations can derive a variety of complex samples.…”

Section: Application Methods Of Multitarget Tracking Algorithm Based ...mentioning

confidence: 99%

Skill Movement Trajectory Recognition of Freestyle Skiing U-Shaped Field Based on Deep Learning and Multitarget Tracking Algorithm

Song

Zhao

2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

Freestyle skiing U-shaped field is a snow sport that uses double boards to perform a series of action skills in a U-shaped pool, which requires very high skills for athletes. In this era of deep learning, in order to develop a more scientific training method, this paper combines multitarget tracking algorithm and deep learning to conduct research in freestyle skiing U-shaped venue skills motion capture. Therefore, this paper combines the convolutional neural network and multitarget tracking algorithm in deep learning to study the human action recognition technology, and then uses the LSTM module to study the freestyle skiing U-shaped venue skills. Finally, this paper designs the training method of the action recognition algorithm of the freestyle U-shaped skiing skills multitarget tracking algorithm based on deep learning. This paper also designs multitarget tracking dataset experiments and model updating experiments. Based on the data of experimental analysis, the training method designed in this paper is optimized, and finally compared with the traditional training method. Compared with the traditional freestyle U-shaped skiing skills training method, the experimental results show that the training method of the freestyle U-shaped skiing skills multitarget tracking algorithm action recognition algorithm is based on deep learning designed in this paper and this improves the skill score by 14.48%. Most professional students are very satisfied with the training method designed in this paper.

show abstract

Section: Application Methods Of Multitarget Tracking Algorithm Based ...mentioning

confidence: 99%

Skill Movement Trajectory Recognition of Freestyle Skiing U-Shaped Field Based on Deep Learning and Multitarget Tracking Algorithm

Song

Zhao

2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

show abstract

“…e phantom constrains the approach, which employs the minimal energy of each point in space to forecast massive deformations and derive the shape of 3D clothing in its equilibrium state. e attire of the virtual phantom has been examined in the literature [18]. e classic is used by the model with proton spring.…”

Section: Related Workmentioning

confidence: 99%

Virtual Garment Piece Design and Stitching Algorithm Based on Virtual Simulation Technology

2022

Security and Communication Networks

View full text Add to dashboard Cite

Modern computer-aided design (CAD) technology is widely used in the field of fashion design. Designers do not have to draw again every time they carry out template design, but can draw and change graphic design drawings directly on the computer using CAD technology. In this way, the designer can see the fitting effect of two-dimensional clothing on the three-dimensional human body in real time in the process of garment graphic design, and the three-dimensional fitting results can change in real time according to the change in two-dimensional clothing pieces. Then, the efficiency of garment design will be greatly improved, and the labor cost and resource time cost will be greatly reduced, and if the cloth of fitting can be highly simulated, the virtual fitting will have more real effect and higher value. Virtual clothing design has always played an important role in the clothing manufacturing industry. In recent years, there are more and more application scenarios of computer-aided design of clothing, and it has good engineering application value to study and realize a reasonable and efficient virtual sewing algorithm for clothing pieces. At home and abroad, the research on 3D clothing fitting technology has gradually deepened, and a series of research results have been achieved in the aspects of clothing piece modeling, clothing piece stitching, and collision detection. However, the traditional collision detection algorithm has an infinite loop in the establishment of the bounding box. At the same time, the time complexity of the collision detection algorithm is high. Based on the “personalized virtual clothing design system platform” independently developed by the research group, this study realizes the interactive design function of clothing pieces and researches and realizes the sewing and folding algorithm of clothing pieces. The clothing piece is triangulated, the generated triangular mesh is used to simulate clothing, finite element analysis is used to simulate the physical movement of clothing, and distance field technology is used to simulate the collision effect between the 3D human model and the clothing piece. To achieve the realistic folding effect of 3D virtual clothes, a seam folding algorithm is suggested, which puts extra bending force limits on patches near the seam, and it works well in the test environment. The stress state of the garment is recreated by assessing the internal and exterior pressures on the fabric. Second, using three-dimensional virtual stitching technology, the two-dimensional clothing is turned into a three-dimensional garment. Finally, the bounding box tree for the clothing and the human body is built. Finally, the bounding box detection algorithm detects and handles the collision between the fabric and the human body during sewing. Finally, texture mapping between fabric and bitmap is used to approximate the garment’s true texture effect. Finally, by investigating the application process of wind force on the fabric, the dynamic impact of cloth under wind is simulated. At present, this technology can simulate the dressing effect of conventional clothing categories, and it can obtain ideal dynamic simulation effect of cloth under the condition of wind blowing.

show abstract

“…Few attempts reported in literature (e.g., [23]) have tried to characterize fabrics via objective means using chemical, physical, or mechanical measurements, but with limited success. We believe texture image analysis is a feasible approach here, as the target characteristics are visual and tactile properties observed at fabric surfaces which have been demonstrated to be generally correlated with visual properties of texture images [24], [25].…”

Section: Role Of Texture In Objects and Scenesmentioning

confidence: 99%

Fabric Surface Characterization: Assessment of Deep Learning-based Texture Representations Using a Challenging Dataset

Long

Sundaresan

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

A High-Speed, Real-Time Vision System for Texture Tracking and Thread Counting

Cited by 11 publications

References 20 publications

Skill Movement Trajectory Recognition of Freestyle Skiing U-Shaped Field Based on Deep Learning and Multitarget Tracking Algorithm

Skill Movement Trajectory Recognition of Freestyle Skiing U-Shaped Field Based on Deep Learning and Multitarget Tracking Algorithm

Virtual Garment Piece Design and Stitching Algorithm Based on Virtual Simulation Technology

Fabric Surface Characterization: Assessment of Deep Learning-based Texture Representations Using a Challenging Dataset

Contact Info

Product

Resources

About