36 research outputs found
The data presented in Figure 6, the 120 mare domes hosting no RMDSs shown in Fig. 5, and all the RMDSs (more than 6400 in number) identified in Mare Tranquillitatis
No full textThe datasets presented here are provided as supporting information for our work, the detail of which is as below:
Feng Zhang, James W. Head, Lionel Wilson, Yibo Meng, Christian Wӧhler, Dijun Guo, Shengli Niu, Roberto Bugiolacchi, Le Qiao, Yanan Dang, Yang Liu, Yongliao Zou
Insights into Lunar Basaltic Volcanism from Mare Domes Superposed by Ring-Moat Dome Structures (RMDSs) in Mare Tranquillitatis
Journal of Geophysical Research: Planets, submitted
Small craters population as a useful geological investigative tool: Apollo 17 region as a case study - Craters Data
No full textCrater coordinates for areas under investigation. Unit names ref. to paper.
Data format: Lat,Lon,Diam_km,Diam_m and Area in Km^2</p
Scatter plots as per Fig. 4, and Spectra-derived TiO2 as per Fig. 7
No full text<p>F. Zhang, J. W. Head, C. Wöhler, R. Bugiolacchi, L. Wilson, A. T. Basilevsky, A. Grumpe, Y. L. Zou.</p>
<p>Ring-Moat Dome Structures (RMDSs) in the Lunar Maria: Statistical, Compositional, and Morphological Characterization and Assessment of Theories of Origin.</p>
<p>Journal of Geophysical Research: Planets, 125, e2019JE005967. https://doi.org/10.1029/2019JE005967</p>
<p> </p>
<p><strong>Figure 4.</strong> Scatter plots of diameters, heights, and volumes of the 532 RMDSs measured from NAC-derived DTMs: (a) diameter vs. height; (b) diameter vs. volume</p>
<p><strong>Figure 7. </strong>M<sup>3</sup>-spectra-derived TiO<sub>2</sub> contents of RMDSs and global maria (latitude range +/- 75 degrees).<strong> </strong>(a) Mean TiO<sub>2</sub> wt% vs. diameter for 2407 RMDSs in seven large mare regions indicated in Figure 2; (b) The TiO<sub>2</sub> content histogram of the measured 2407 RMDSs (green color), compared with the global TiO<sub>2</sub> distribution of all lunar mare regions (red color). Both histograms are normalized to 1, such that the sum of all red or green histogram bins is 1, respectively. The global Ti abundance was taken from the complete lunar mare surface at a latitude range +/- 75 degrees</p>
<p> </p>
The crater detection results (necessary data for making Figs 6 and 7) for areas A2, A4 and A7 to estimate the ages of RMDS-overlapped craters in Figs 2a-2c
No full textTo estimate the ages of RMDS-overlapped craters in Figs 2a-2c ( located in areas A2, A4 and A7, respectively) with a locally calibrated Monte Carlo model, the NAC-DEM-based crater detection results are provided here, as the basis input and necessary data to make Figs 6 and 7.
They have been introduced in the following publication, where more details can be found:
F. Zhang, J. W. Head, C. Wöhler, A. T. Basilevsky, L. Wilson, M. Xie, R. Bugiolacchi, T. Wilhelm, S. Althoff, Y. L. Zou.
The Lunar Mare Ring-Moat Dome Structure (RMDS) Age Conundrum: Contemporaneous with Imbrian-Aged Host Lava Flows or Emplaced in the Copernican?
Journal of Geophysical Research: Planets, 126, e2021JE006880. https://doi. org/10.1029/2021JE00688
CE-2_Ch4-Ch1_midnight_MRM
No full textTemperature Brightness differences (TB∆1-4) at midnight between channels 1 (3 GHz) and 4 (37 GHz) from the Chang’E-2 Microwave Radiometers (MRMs) data
Improving generative adversarial network inversion via fine-tuning GAN encoders
No full textGenerative adversarial networks (GANs) can synthesize high-quality (HQ) images, and GAN inversion is a technique that discovers how to invert given images back to latent space. While existing methods perform on StyleGAN inversion, they have limited performance and are not generalized to different GANs. To address these issues, we proposed a self-supervised method to pre-train and fine-tune GAN encoders. First, we designed an adaptive block to fit different encoder architectures for inverting diverse GANs. Then we pre-train GAN encoders using synthesized images and emphasize local regions through cropping images. Finally, we fine-tune the pre-trained GAN encoder for inverting real images. Compared with state-of-the-art methods, our method achieved better results that reconstructed high-quality images on mainstream GANs. Our code and pre-trained models are available at: https://github.com/disanda/Deep-GAN-Encoders
NIR spectral investigation of the Delisle/Diophantus crater region by the SIR-2 instrument
No full text2019-Xunyu Zhang-Mafic minerals in the South Pole-Aitken basin
No full textIt is the dataset of extracted spectra and their location used in the paper "Mafic minerals in the South Pole-Aitken basin".</p
Fast 2-Step Regularization on Style Optimization for Real Face Morphing
No full textStyleGAN is now capable of achieving excellent results, especially high-quality face synthesis. Recently, some studies have tried to invert real face images into style latent space via StyleGAN. However, morphing real faces via latent representation is still in its infancy. Training costs are high and/or require huge samples with labels. By adding regularization to style optimization, we propose a novel method to morph real faces based on StyleGAN. To do the supervised task, we label latent vectors via synthesized faces and release the label set; then we utilise logistic regression to fast discover interpretable directions in latent space. Appropriate regularization helps us to optimize both latent vectors (faces and directions). Moreover, we use learned directions under different layer representations to handle real face morphing. Compared to the existing methods, our method faster yields a larger diverse and realistic output. Code and cases are available at \url{https://github.com/disanda/RFM}.</p
