scholarly journals Deep Representation of a Normal Map for Screen-Space Fluid Rendering

2021 ◽  
Vol 11 (19) ◽  
pp. 9065
Author(s):  
Myungjin Choi ◽  
Jee-Hyeok Park ◽  
Qimeng Zhang ◽  
Byeung-Sun Hong ◽  
Chang-Hun Kim
Keyword(s):  
Surface Reconstruction ◽  
Computation Time ◽  
Fluid Simulation ◽  
Image Resolution ◽  
Reconstruction Method ◽  
Generative Adversarial Network ◽  
Normal Map ◽  
Adversarial Network ◽  
Reconstruction Methods ◽  
Screen Space

We propose a novel method for addressing the problem of efficiently generating a highly refined normal map for screen-space fluid rendering. Because the process of filtering the normal map is crucially important to ensure the quality of the final screen-space fluid rendering, we employ a conditional generative adversarial network (cGAN) as a filter that learns a deep normal map representation, thereby refining the low-quality normal map. In particular, we have designed a novel loss function dedicated to refining the normal map information, and we use a specific set of auxiliary features to train the cGAN generator to learn features that are more robust with respect to edge details. Additionally, we constructed a dataset of six different typical scenes to enable effective demonstrations of multitype fluid simulation. Experiments indicated that our generator was able to infer clearer and more detailed features for this dataset than a basic screen-space fluid rendering method. Moreover, in some cases, the results generated by our method were even smoother than those generated by the conventional surface reconstruction method. Our method improves the fluid rendering results via the high-quality normal map while preserving the advantages of the screen-space fluid rendering methods and the traditional surface reconstruction methods, including that of the computation time being independent of the number of simulation particles and the spatial resolution being related only to image resolution.

scholarly journals Classification of Space Objects by Using Deep Learning with Micro-Doppler Signature Images

Sensors ◽  
2021 ◽  
Vol 21 (13) ◽  
pp. 4365
Author(s):  
Kwangyong Jung ◽  
Jae-In Lee ◽  
Nammoon Kim ◽  
Sunjin Oh ◽  
Dong-Wook Seo
Keyword(s):  
Transfer Learning ◽  
Data Augmentation ◽  
Doppler Frequency ◽  
Extraction Methods ◽  
Classification Performance ◽  
Reconstruction Method ◽  
Generative Adversarial Network ◽  
Adversarial Network ◽  
Space Objects ◽  
Target Characteristics

Radar target classification is an important task in the missile defense system. State-of-the-art studies using micro-doppler frequency have been conducted to classify the space object targets. However, existing studies rely highly on feature extraction methods. Therefore, the generalization performance of the classifier is limited and there is room for improvement. Recently, to improve the classification performance, the popular approaches are to build a convolutional neural network (CNN) architecture with the help of transfer learning and use the generative adversarial network (GAN) to increase the training datasets. However, these methods still have drawbacks. First, they use only one feature to train the network. Therefore, the existing methods cannot guarantee that the classifier learns more robust target characteristics. Second, it is difficult to obtain large amounts of data that accurately mimic real-world target features by performing data augmentation via GAN instead of simulation. To mitigate the above problem, we propose a transfer learning-based parallel network with the spectrogram and the cadence velocity diagram (CVD) as the inputs. In addition, we obtain an EM simulation-based dataset. The radar-received signal is simulated according to a variety of dynamics using the concept of shooting and bouncing rays with relative aspect angles rather than the scattering center reconstruction method. Our proposed model is evaluated on our generated dataset. The proposed method achieved about 0.01 to 0.39% higher accuracy than the pre-trained networks with a single input feature.

Sound Field Reconstruction in Rooms with Deep Generative Models

2021 ◽  
Vol 263 (5) ◽  
pp. 1527-1538
Author(s):  
Xenofon Karakonstantis ◽  
Efren Fernandez Grande
Keyword(s):  
Plane Waves ◽  
Sound Field ◽  
Real Data ◽  
Generative Models ◽  
Random Wave ◽  
Generative Adversarial Network ◽  
Underlying Distribution ◽  
Adversarial Network ◽  
Reconstruction Methods ◽  
Free Region

The characterization of Room Impulse Responses (RIR) over an extended region in a room by means of measurements requires dense spatial with many microphones. This can often become intractable and time consuming in practice. Well established reconstruction methods such as plane wave regression show that the sound field in a room can be reconstructed from sparsely distributed measurements. However, these reconstructions usually rely on assuming physical sparsity (i.e. few waves compose the sound field) or trait in the measured sound field, making the models less generalizable and problem specific. In this paper we introduce a method to reconstruct a sound field in an enclosure with the use of a Generative Adversarial Network (GAN), which s new variants of the data distributions that it is trained upon. The goal of the proposed GAN model is to estimate the underlying distribution of plane waves in any source free region, and map these distributions from a stochastic, latent representation. A GAN is trained on a large number of synthesized sound fields represented by a random wave field and then tested on both simulated and real data sets, of lightly damped and reverberant rooms.

scholarly journals Promising Generative Adversarial Network Based Sinogram Inpainting Method for Ultra-Limited-Angle Computed Tomography Imaging

Sensors ◽  
2019 ◽  
Vol 19 (18) ◽  
pp. 3941 ◽  
Author(s):  
Li ◽  
Cai ◽  
Wang ◽  
Zhang ◽  
Tang ◽  
...  
Keyword(s):  
Computed Tomography ◽  
Imaging System ◽  
Reconstruction Method ◽  
Ct Reconstruction ◽  
Reconstruction Algorithms ◽  
Generative Adversarial Network ◽  
Image Domain ◽  
Adversarial Network ◽  
Domain Loss ◽  
Limited Angle

Limited-angle computed tomography (CT) image reconstruction is a challenging problem in the field of CT imaging. In some special applications, limited by the geometric space and mechanical structure of the imaging system, projections can only be collected with a scanning range of less than 90°. We call this kind of serious limited-angle problem the ultra-limited-angle problem, which is difficult to effectively alleviate by traditional iterative reconstruction algorithms. With the development of deep learning, the generative adversarial network (GAN) performs well in image inpainting tasks and can add effective image information to restore missing parts of an image. In this study, given the characteristic of GAN to generate missing information, the sinogram-inpainting-GAN (SI-GAN) is proposed to restore missing sinogram data to suppress the singularity of the truncated sinogram for ultra-limited-angle reconstruction. We propose the U-Net generator and patch-design discriminator in SI-GAN to make the network suitable for standard medical CT images. Furthermore, we propose a joint projection domain and image domain loss function, in which the weighted image domain loss can be added by the back-projection operation. Then, by inputting a paired limited-angle/180° sinogram into the network for training, we can obtain the trained model, which has extracted the continuity feature of sinogram data. Finally, the classic CT reconstruction method is used to reconstruct the images after obtaining the estimated sinograms. The simulation studies and actual data experiments indicate that the proposed method performed well to reduce the serious artifacts caused by ultra-limited-angle scanning.

scholarly journals Low Resolution Face Recognition Using Generative Adversarial Network (GAN)

2020 ◽  
Author(s):  
Howard Martin ◽  
Suharjito
Keyword(s):  
Face Recognition ◽  
Super Resolution ◽  
Image Data ◽  
Face Image ◽  
Image Resolution ◽  
Low Resolution ◽  
Generative Adversarial Network ◽  
Resolution Image ◽  
Surveillance Camera ◽  
Adversarial Network

Abstract Face recognition has a lot of use on smartphone authentication, finding people, etc. Nowadays, face recognition with a constrained environment has achieved very good performance on accuracy. However, the accuracy of existing face recognition methods will gradually decrease when using a dataset with an unconstrained environment. Face image with an unconstrained environment is usually taken from a surveillance camera. In general, surveillance cameras will be placed on the corner of a room or even on the street. So, the image resolution will be low. Low-resolution image will cause the face very hard to be recognized and the accuracy will eventually decrease. That is the main reason why increasing the accuracy of the Low-Resolution Face Recognition (LRFR) problem is still challenging. This research aimed to solve the Low-Resolution Face Recognition (LRFR) problem. The datasets are YouTube Faces Database (YTF) and Labelled Faces in The Wild (LFW). In this research, face image resolution would be decreased using bicubic linear and became the low-resolution image data. Then super resolution methods as the preprocessing step would increase the image resolution. Super resolution methods used in this research are Super resolution GAN (SRGAN) [1] and Enhanced Super resolution GAN (ESRGAN) [2]. These methods would be compared to reach a better accuracy on solving LRFR problem. After increased the image resolution, the image would be recognized using FaceNet. This research concluded that using super resolution as the preprocessing step for LRFR problem has achieved a higher accuracy compared to [3]. The highest accuracy achieved by using ESRGAN as the preprocessing and FaceNet for face recognition with accuracy of 98.96 % and Validation rate 96.757 %.

Sign in / Sign up

Export Citation Format

Share Document