检索结果-内蒙古大学图书馆

Determining intrinsic optical properties of 3D printing materials

Optics Express 2025年第13期33卷 28493-28510页

作者： Lubna Abu Rmaileh Philipp Nguyen Alexer Kissel Philipp Urban Fraunhofer Institute for Computer Graphics Research IGD Darmstadt Germany Norwegian University of Science and Technology NTNU Gjøvik Norway Institute for Laser Technologies in Medicine and Metrology Ulm Germany

Accurate knowledge of the intrinsic optical properties of 3D printing materials, i.e., spectral absorption and scattering coefficients, phase function, and refractive index, is essential for simulating the appearance of translucent prints on displays (softproofing) or optimizing material arrangements to achieve desired optical effects in multi-material 3D prints. This information is also critical for designing printing materials that mimic the optical characteristics of other materials, a key requirement in applications like dental restorations. Current methods for measuring these properties rely on specialized laboratory equipment and expert knowledge. In this paper, we propose an approach that uses a commercial reflectance/transmittance spectrophotometer to determine the spectral absorption and scattering coefficients and refractive index of 3D printing materials. We model the light path of this device to simulate reflectance and transmittance measurements via a Monte Carlo path tracer. We then predict measurements for a large set of random but plausible intrinsic optical material properties for three different sample thicknesses. With these data, we train machine learning models to infer the intrinsic properties from phenomenological reflectance/transmittance measurements, considering a priori knowledge of smoothness as a regularization constraint. We validate our method by comparing results for real printing materials with accurate laboratory measurements and provide the trained machine learning models to the community.

关键词： 3D printing Absorption coefficient Material properties Optical materials Optical properties Stochastic gradient descent

来源：评论

学校读者我要写书评

暂无评论

Automated classification of crests on pottery sherds using pattern recognition on 2D images

Automated classification of crests on pottery sherds using p...

引用

20th Eurographics Workshop on graphics and Cultural Heritage, GCH 2022

作者： Ritz, M. Santos, P. Fellner, D.W. Fraunhofer IGD Germany TU-Darmstadt Germany Graz University of Technology Institute of Computer Graphics and Knowledge Visualization Austria

ISBN: (纸本)9783038681786

Manual classification of artefacts is a labor intensive process. Based on 2D images and 3D scans of - for example - ceramic shards, we developed a pattern recognition algorithm which automatically extracts relief features for each newly recorded object and tries to automate the classification process. Based on characteristics found, previously unknown objects are automatically corelated to already classified objects of a collection exhibiting the greatest similarity. As a result, classes of artefacts form iteratively, which ultimately also corresponds to the overall goal which is the automated classification of entire collections. The greatest challenge in developing our software approach was the heterogeneity of reliefs, and in particular the fact that current machine learning approaches were out of question due to the very limited number of objects per class. This led to the implementation of an analytical approach that is capable of performing a classification based on very few artefacts. © ISVR,2023 *** rights reserved

关键词： Iterative methods

来源：评论

学校读者我要写书评

暂无评论

GraMMaCAD: INTERACTIVELY DEFINING SPATIALLY VARYING FGMS ON BREP CAD MODELS

GraMMaCAD: INTERACTIVELY DEFINING SPATIALLY VARYING FGMS ON ...

引用

ASME 2022 International Design Engineering Technical Conferences and computers and Information in Engineering Conference, IDETC-CIE 2022

作者： Luu, Thu Huong Altenhofen, Christian Stork, André Fellner, Dieter TU Darmstadt Interactive Graphics Systems Group Darmstadt64283 Germany Fraunhofer IGD Darmstadt64283 Germany Graz University of Technology Institute of Computer Graphics and Knowledge Visualization Graz8010 Austria

ISBN: (纸本)9780791886212

Additive manufacturing (AM) has enabled control over heterogeneous materials and structures in ways that were not previously possible, including functionally graded materials (FGM). However, typical computer aided design (CAD) systems – mainly based on boundary representations (B-Rep) – are not built for modeling and representing spatially varying material information, which are the key features for designing FGM parts. While volumetric representations (V-Rep) such as tetrahedral or hexahedral meshes, trivariate splines or subdivision solids can hold the necessary FGM information, they are by default not compatible with existing CAD models. Thus, the capabilities of AM motivate enhancing CAD systems to support volumetric material allocation and distribution. We present an interactive, intuitive and efficient interaction concept based on grading features and transition functions for augmenting CAD models with smooth volumetric material gradients for multi-material fabrication using our system called GraMMaCAD (Graded Multi-Material CAD). This concept has the advantage of being extremely versatile, since the material property functions are left to the arbitrariness of the user, who is free to modify their definition as needed. By providing interactive preview modes, the system allows for efficient exploration of the volumetric gradation of the designed FGM part. We fabricated two functional examples that demonstrate the usability of the system and show the ability to quickly augment models with functionally graded material transitions. Copyright © 2022 by ASME.

关键词： Functionally graded materials

来源：评论

学校读者我要写书评

暂无评论

Surgical Phase Recognition for different hospitals

引用

Current Directions in Biomedical Engineering 2023年第1期9卷 315-318页

作者： Wisotzky, Eric L. Beckmann, Sophie Eisert, Peter Renz-Kiefel, Lasse Hilsmann, Anna Lünse, Sebastian Mantke, René Fraunhofer Heinrich-Hertz-Institute HHI & Humboldt-University Berlin Germany Computer Vision & Graphics Fraunhofer Heinrich-Hertz-Institute HHI Berlin Germany Brandenburg Medical School Department of Surgery University Hospital Brandenburg Germany Faculty of Health Sciences Joint Faculty of the Brandenburg University of Technology Cottbus - Senftenberg Brandenburg Medical School Theodor Fontane Neuruppin Germany University of Potsdam & Department of Surgery University Hospital Brandenburg Germany

Surgical phase recognition is an important aspect of surgical workflow analysis, as it allows an automatic analysis of the performance and efficiency of surgical procedures. A big challenge for training a neural network for surgical phase recognition is the availability of training data and the large (visual) variability in procedures of different surgeons. Hence, a network must be able to generalize to new data. In this paper, we present an adaptation of a Temporal Convolutional Network for surgical phase recognition in order to ensure the generalization of the network to new scenes with different conditions on the example of cholecystectomy. We used publicly available datasets of 104 surgeries from four different centers for training. The results showed that the network was able to generalize to new scenes and we obtained recognition results with accuracy up to 82% on our own six captured surgeries, performed in a different hospital. This performance is similar for test data from the hospitals of the training data, suggesting that the network can well generalize to new surgical rooms and surgeons. The findings have important implications for the development of automated surgical decision support systems that can be applied in a variety of real-world surgical settings. © 2023 the author(s), published by Walter de Gruyter Berlin/Boston.

关键词： Hospitals

来源：评论

学校读者我要写书评

暂无评论

MIPGAN-Generating Strong and High Quality Morphing Attacks Using Identity Prior Driven GAN

IEEE Transactions on Biometrics, Behavior, and Identity Scie...

引用

IEEE Transactions on Biometrics, Behavior, and Identity Science 2021年第3期3卷 365-383页

作者： Zhang, Haoyu Venkatesh, Sushma Ramachandra, Raghavendra Raja, Kiran Damer, Naser Busch, Christoph Norwegian University of Science and Technology Gjovik2815 Norway Department of Igd Fraunhofer Institute for Computer Graphics Research Igd Darmstadt64283 Germany

Face morphing attacks target to circumvent Face Recognition Systems (FRS) by employing face images derived from multiple data subjects (e.g., accomplices and malicious actors). Morphed images can be verified against contributing data subjects with a reasonable success rate, given they have a high degree of facial resemblance. The success of morphing attacks is directly dependent on the quality of the generated morph images. We present a new approach for generating strong attacks extending our earlier framework for generating face morphs. We present a new approach using an Identity Prior Driven Generative Adversarial Network, which we refer to as MIPGAN (Morphing through Identity Prior driven GAN). The proposed MIPGAN is derived from the StyleGAN with a newly formulated loss function exploiting perceptual quality and identity factor to generate a high quality morphed facial image with minimal artefacts and with high resolution. We demonstrate the proposed approach's applicability to generate strong morphing attacks by evaluating its vulnerability against both commercial and deep learning based Face Recognition System (FRS) and demonstrate the success rate of attacks. Extensive experiments are carried out to assess the FRS's vulnerability against the proposed morphed face generation technique on three types of data such as digital images, re-digitized (printed and scanned) images, and compressed images after re-digitization from newly generated MIPGAN Face Morph Dataset. The obtained results demonstrate that the proposed approach of morph generation poses a high threat to FRS. © 2019 IEEE.

关键词： Face recognition

来源：评论

学校读者我要写书评

暂无评论

If It’s Not Enough, Make It So: Reducing Authentic Data Demand in Face Recognition through Synthetic Faces

arXiv

引用

arXiv 2024年

作者： Atzori, Andrea Boutros, Fadi Damer, Naser Fenu, Gianni Marras, Mirko Department of Mathematics and Computer Science University of Cagliari Cagliari Italy Fraunhofer Institute for Computer Graphics Research Germany Department of Computer Science TU Darmstadt Darmstadt Germany

Recent advances in deep face recognition have spurred a growing demand for large, diverse, and manually annotated face datasets. Acquiring authentic, high-quality data for face recognition has proven to be a challenge, primarily due to privacy concerns. Large face datasets are primarily sourced from web-based images, lacking explicit user consent. In this paper, we examine whether and how synthetic face data can be used to train effective face recognition models with reduced reliance on authentic images, thereby mitigating data collection concerns. First, we explored the performance gap among recent state-of-the-art face recognition models, trained with synthetic data only and authentic (scarce) data only. Then, we deepened our analysis by training a state-of-the-art backbone with various combinations of synthetic and authentic data, gaining insights into optimizing the limited use of the latter for verification accuracy. Finally, we assessed the effectiveness of data augmentation approaches on synthetic and authentic data, with the same goal in mind. Our results highlighted the effectiveness of FR trained on combined datasets, particularly when combined with appropriate augmentation techniques. © 2024, CC BY.

关键词： Face recognition

来源：评论

学校读者我要写书评

暂无评论

Unveiling the Two-Faced Truth: Disentangling Morphed Identities for Face Morphing Detection

Unveiling the Two-Faced Truth: Disentangling Morphed Identit...

引用

European Signal Processing Conference (EUSIPCO)

作者： Eduarda Caldeira Pedro C. Neto Tiago Gonçalves Naser Damer Ana F. Sequeira Jaime S. Cardoso INESC TEC University of Porto TU Darmstadt Fraunhofer Institute for Computer Graphics Research IGD

Morphing attacks keep threatening biometric systems, especially face recognition systems. Over time they have become simpler to perform and more realistic, as such, the usage of deep learning systems to detect these attacks has grown. At the same time, there is a constant concern regarding the lack of interpretability of deep learning models. Balancing performance and interpretability has been a difficult task for scientists. However, by leveraging domain information and proving some constraints, we have been able to develop IDistill, an interpretable method with state-of-the-art performance that provides information on both the identity separation on morph samples and their contribution to the final prediction. The domain information is learnt by an autoencoder and distilled to a classifier system in order to teach it to separate identity information. When compared to other methods in the literature it outperforms them in three out of five databases and is competitive in the remaining.

关键词：

来源：评论

学校读者我要写书评

暂无评论

If It's Not Enough, Make It So: Reducing Authentic Data Demand in Face Recognition through Synthetic Faces

If It's Not Enough, Make It So: Reducing Authentic Data Dema...

引用

International Conference on Automatic Face and Gesture Recognition

作者： Andrea Atzori Fadi Boutros Naser Damer Gianni Fenu Mirko Marras Department of Mathematics and Computer Science University of Cagliari Cagliari Italy Fraunhofer Institute for Computer Graphics Research Germany Department of Computer Science TU Darmstadt Darmstadt Germany

ISBN: (数字)9798350394948

ISBN: (纸本)9798350394955

Recent advances in deep face recognition have spurred a growing demand for large, diverse, and manually annotated face datasets. Acquiring authentic, high-quality data for face recognition has proven to be a challenge, primarily due to privacy concerns. Large face datasets are primarily sourced from web-based images, lacking explicit user consent. In this paper, we examine whether and how synthetic face data can be used to train effective face recognition models with reduced reliance on authentic images, and thus mitigating large authentic data collection concerns. First, we explored the performance gap among recent state-of-the-art face recognition models, trained only on synthetic data or authentic data. Then, we deepened our analysis by training a state-of-the-art back-bone with various combinations of synthetic and authentic data, gaining insights into optimizing the limited use of the latter for verification accuracy. Finally, we assessed the effectiveness of data augmentation approaches on synthetic and authentic data, with the same goal in mind. Our results highlighted the effectiveness of FR trained on combined datasets, particularly when combined with appropriate augmentation techniques.

关键词： Training Data privacy Accuracy Face recognition Gesture recognition Data collection Data augmentation

来源：评论

学校读者我要写书评

暂无评论

AdaDistill: Adaptive Knowledge Distillation for Deep Face Recognition

arXiv

引用

arXiv 2024年

作者： Boutros, Fadi Štruc, Vitomir Damer, Naser Fraunhofer Institute for Computer Graphics Research IGD Darmstadt Germany University of Ljubljana Ljubljana Slovenia Department of Computer Science TU Darmstadt Darmstadt Germany

Knowledge distillation (KD) aims at improving the performance of a compact student model by distilling the knowledge from a high-performing teacher model. In this paper, we present an adaptive KD approach, namely AdaDistill, for deep face recognition. The proposed AdaDistill embeds the KD concept into the softmax loss by training the student using a margin penalty softmax loss with distilled class centers from the teacher. Being aware of the relatively low capacity of the compact student model, we propose to distill less complex knowledge at an early stage of training and more complex one at a later stage of training. This relative adjustment of the distilled knowledge is controlled by the progression of the learning capability of the student over the training iterations without the need to tune any hyper-parameters. Extensive experiments and ablation studies show that AdaDistill can enhance the discriminative learning capability of the student and demonstrate superiority over various state-of-the-art competitors on several challenging benchmarks, such as IJB-B, IJB-C, and ICCV2021-MFR. © 2024, CC BY-NC-SA.

关键词： Distillation

来源：评论

学校读者我要写书评

暂无评论

Bias and Diversity in Synthetic-based Face Recognition

arXiv

引用

arXiv 2023年

作者： Huber, Marco Luu, Anh Thi Boutros, Fadi Kuijper, Arjan Damer, Naser Fraunhofer Institute for Computer Graphics Research IGD Darmstadt Germany Department of Computer Science TU Darmstadt Darmstadt Germany

Synthetic data is emerging as a substitute for authentic data to solve ethical and legal challenges in handling authentic face data. The current models can create real-looking face images of people who do not exist. However, it is a known and sensitive problem that face recognition systems are susceptible to bias, i.e. performance differences between different demographic and non-demographics attributes, which can lead to unfair decisions. In this work, we investigate how the diversity of synthetic face recognition datasets compares to authentic datasets, and how the distribution of the training data of the generative models affects the distribution of the synthetic data. To do this, we looked at the distribution of gender, ethnicity, age, and head position. Furthermore, we investigated the concrete bias of three recent synthetic-based face recognition models on the studied attributes in comparison to a baseline model trained on authentic data. Our results show that the generator generate a similar distribution as the used training data in terms of the different attributes. With regard to bias, it can be seen that the synthetic-based models share a similar bias behavior with the authentic-based models. However, with the uncovered lower intra-identity attribute consistency seems to be beneficial in reducing bias. Copyright © 2023, The Authors. All rights reserved.

关键词： Face recognition

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：