Filters








11,731 Hits in 6.3 sec

Focal Frequency Loss for Image Reconstruction and Synthesis [article]

Liming Jiang, Bo Dai, Wayne Wu, Chen Change Loy
2021 arXiv   pre-print
In this study, we show that narrowing gaps in the frequency domain can ameliorate image reconstruction and synthesis quality further.  ...  Image reconstruction and synthesis have witnessed remarkable progress thanks to the development of generative models.  ...  This study is supported under the RIE2020 Industry Alignment Fund -Industry Collaboration Projects (IAF-ICP) Funding Initiative, as well as cash and in-kind contribution from the industry partner(s).  ... 
arXiv:2012.12821v3 fatcat:tvtmi2fag5fpnjqbuaesy2p4kq

A Unified Learning Based Framework for Light Field Reconstruction from Coded Projections [article]

Anil Kumar Vadathya, Sharath Girish, Kaushik Mitra
2019 arXiv   pre-print
Especially, our LF reconstructions from focus-defocus pair is comparable to other learning-based view synthesis approaches from multiple images.  ...  Our algorithm consists of three stages 1) we recover the all-in-focus image from the coded image 2) we estimate the disparity maps for all the LF views from the coded image and the all-in-focus image,  ...  To compensate for the loss of angular information in the defocused image, we propose to use the all-in-focus image as well for LF reconstruction.  ... 
arXiv:1812.10532v2 fatcat:ots4sovzkrgjrae262p5qyau4m

Deep Anti-aliasing of Whole Focal Stack Using Slice Spectrum [article]

Yaning Li, Xue Wang, Hao Zhu, Guoqing Zhou, Qing Wang
2021 arXiv   pre-print
We first explore the structural characteristics embedded in the focal stack slice and its corresponding frequency-domain representation, i.e., the Focal Stack Spectrum (FSS).  ...  Based on these two observations, we propose a learning-based FSS reconstruction approach for one-time aliasing removing over the whole focal stack.  ...  The loss function is, loss = F − F gt 2 + λ • loss s , (10) where loss s constrains the conjugate symmetry of the reconstructed FSS.  ... 
arXiv:2101.09420v2 fatcat:hf36o5mxona7dgxs54xdu336vi

SAMURAI: Shape And Material from Unconstrained Real-world Arbitrary Image collections [article]

Mark Boss, Andreas Engelhardt, Abhishek Kar, Yuanzhen Li, Deqing Sun, Jonathan T. Barron, Hendrik P. A. Lensch, Varun Jampani
2022 arXiv   pre-print
Our method works on in-the-wild online image collections of an object and produces relightable 3D assets for several use-cases such as AR/VR.  ...  We propose a joint optimization framework to estimate the shape, BRDF, and per-image camera pose and illumination.  ...  Acknowledgments and Disclosure of Funding  ... 
arXiv:2205.15768v1 fatcat:jssi6ry4rjeknefwr6q7jd3eyy

Optimal tilt magnitude determination for aberration-corrected super resolution exit wave function reconstruction

S. J. Haigh, H. Sawada, A. I. Kirkland
2009 Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences  
We have determined the optimal experimental tilt magnitude for aperture synthesis in an aberration-corrected TEM.  ...  Transmission electron microscope (TEM) images recorded under tilted illumination conditions transfer higher spatial frequencies than axial images.  ...  Hetherington and JEOL Ltd for technical assistance and Dr C. Kieselowski for the provision of the gold sample used. Financial support from EPSRC (Grant EP/F048009/1) is gratefully acknowledged.  ... 
doi:10.1098/rsta.2009.0124 pmid:19687064 fatcat:3fm2pkikanbrbfbe7ajf25buhi

Three-dimensional imaging of single isolated cell nuclei using optical projection tomography

Mark Fauver, Eric J. Seibel, J. Richard Rahn, Michael G. Meyer, Florence W. Patten, Thomas Neumann, Alan C. Nelson
2005 Optics Express  
The resolution is found to be 0.9 µm for both extended DOF images and 3D reconstructed images.  ...  To assess loss of resolution from fixed focus image to extended DOF image to 3D reconstructed image, the 10-90% rise distance is measured for a dyed microsphere.  ...  Thanks to Jeff Magula for assistance in design and fabrication of the microcapillary tubebased rotation stage.  ... 
doi:10.1364/opex.13.004210 pmid:19495335 fatcat:s33wh7nalvhy3n6miklnhvo444

Deep 3D-Zoom Net: Unsupervised Learning of Photo-Realistic 3D-Zoom [article]

Juan Luis Gonzalez Bello, Munchurl Kim
2019 arXiv   pre-print
In contrast, the optical zoom changes the focal length and the digital zoom is used to enlarge a certain region of an image to the original image size.  ...  network architecture that models depth-image-based rendering (DIBR), taking into account high-frequency details without the need for estimating the intermediate disparity; and (iii) incorporating a discriminator  ...  loss Our reconstruction loss is defined as a combination of two terms, appearance loss and perceptual loss, as l rec = 0.8l ap + 0.2l p (15) Appearance loss.  ... 
arXiv:1909.09349v2 fatcat:wipuhls5fjexbj7tpftnz7yrp4

Page 794 of SMPTE Motion Imaging Journal Vol. 75, Issue 8 [page]

1966 SMPTE Motion Imaging Journal  
. + Fp*Fo The Ramachandran and Raman a- synthesis reconstructs the crystal structure by taking the Fourier transform of Fp|F |’, that is, Fp|Fp|? + Fp Fe|?  ...  In the centro- symmetric crystal structure the two over- lapping images coincide. A second-order improvement on the a-synthesis is produced by the 8-synthesis* in which |Fy|?/Fp* replaces Fp-|Fy|?  ... 

Experimental verification of sub-wavelength focusing via a holographic metallic screen

A. Wong, G.V. Eleftheriades
2008 2008 IEEE Antennas and Propagation Society International Symposium  
for screen synthesis.  ...  It can be seen that within a small range of frequencies around 10GHz, the change in frequency amounts to a tradeoff between the focal width and side lobe levels.  ... 
doi:10.1109/aps.2008.4619388 fatcat:kslc5jz35ncsjiijtdmemlhdhq

WaveGAN: Frequency-aware GAN for High-Fidelity Few-shot Image Generation [article]

Mengping Yang, Zhe Wang, Ziqiu Chi, Wenyi Feng
2022 arXiv   pre-print
Moreover, we utilize a frequency L1-loss on the generated and real images to further impede frequency information loss.  ...  To address this, we propose WaveGAN, a frequency-aware model for few-shot image generation.  ...  For example, Jiang et al. propose focal frequency loss to avoid the loss of important frequency information for image reconstruction tasks [18] .  ... 
arXiv:2207.07288v2 fatcat:6r7c6pmlsnhzzm5y6jyqoo45mm

Perception-Enhanced Image Super-Resolution via Relativistic Generative Adversarial Networks [chapter]

Thang Vu, Tung M. Luu, Chang D. Yoo
2019 Lecture Notes in Computer Science  
perceptual quality of the reconstructed images by considering the following three issues: (1) ease GAN training by replacing an absolute with a relativistic discriminator, (2) include in the loss function  ...  This paper considers a deep Generative Adversarial Networks (GAN) based method referred to as the Perception-Enhanced Super-Resolution (PESR) for Single Image Super Resolution (SISR) that enhances the  ...  High-frequency noise amplification is inevitable with GAN based synthesis, and in order to mitigate this problem, the total variance loss function [43] is considered.  ... 
doi:10.1007/978-3-030-11021-5_7 fatcat:v6mizuzxzrctnoize3xcpf7d4i

A Full-Aperture Image Synthesis Method for the Rotating Rectangular Aperture System Using Fourier Spectrum Restoration

Guomian Lv, Hao Xu, Huajun Feng, Zhihai Xu, Hao Zhou, Qi Li, Yueting Chen
2021 Photonics  
In this paper, we propose a novel full-aperture image synthesis algorithm for the RRA system, based on Fourier spectrum restoration.  ...  Its imaging quality depends largely on the image synthesis algorithm, and the mainstream multi-frame deblurring approach is sophisticated and time-consuming.  ...  Acknowledgments: We thank Bian meijuan from the facility platform of optical engineering of Zhejiang University for instrument support.  ... 
doi:10.3390/photonics8110522 fatcat:4xmrs4k7ufgbbhuewachovsoii

Tomographic near-eye displays

Seungjae Lee, Youngjin Jo, Dongheon Yoo, Jaebum Cho, Dukho Lee, Byoungho Lee
2019 Nature Communications  
The ultimate 3D displays should provide both psychological and physiological cues for depth recognition.  ...  We conclude with a detailed analysis and thorough discussion of the display's optimal volumetric reconstruction. of tomographic displays.  ...  The DMD and the HDR LED are synchronized for direct digital synthesis 26 that decomposes color images into binary image sequences.  ... 
doi:10.1038/s41467-019-10451-2 pmid:31175279 pmcid:PMC6555831 fatcat:6umjkhd46zb7dmd3f4dfysqzu4

SiCloPe: Silhouette-Based Clothed People [article]

Ryota Natsume, Shunsuke Saito, Zeng Huang, Weikai Chen, Chongyang Ma, Hao Li, Shigeo Morishima
2019 arXiv   pre-print
Our method can reconstruct a complete and textured 3D model of a person wearing clothes from a single input picture.  ...  Our experiments demonstrate that our silhouette-based model is an effective representation and the appearance of the back view can be predicted reliably using an image-to-image translation network.  ...  For the front-to-back image synthesis, we set the yaw angle to be frontal and sample the pitch angle from 0, 7.5, 15 • with a focal length of 800mm.  ... 
arXiv:1901.00049v2 fatcat:2orw6o6l5vfzpivggdtxuo7aeu

SiCloPe: Silhouette-Based Clothed People

Ryota Natsume, Shunsuke Saito, Zeng Huang, Weikai Chen, Chongyang Ma, Hao Li, Shigeo Morishima
2019 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)  
Our method can reconstruct a complete and textured 3D model of a person wearing clothes from a single input picture.  ...  Our experiments demonstrate that our silhouette-based model is an effective representation and the appearance of the back view can be predicted reliably using an image-to-image translation network.  ...  Hao Li is affiliated with the University of Southern California, the USC Institute for Creative Technologies, and Pinscreen.  ... 
doi:10.1109/cvpr.2019.00461 dblp:conf/cvpr/NatsumeSH0MLM19 fatcat:fmmii7wry5gcvbvmo3xdbevuya
« Previous Showing results 1 — 15 out of 11,731 results