id
stringlengths 20
32
| text
stringlengths 100
121k
| image_path
stringlengths 15
235
|
|---|---|---|
arxivcap_2202.00198_3 | Title: Recognition-Aware Learned Image Compression | Caption: The joint compression-recognition architecture, where the encoders, decoders, Gaussian Mixture Model (GMM), and factorized entropy model are as in <cit.>. The recognition network is an EfficientNet-B0 as in <cit.>. $x$ is an input image, $y$ are the latents, $z$ are the hyper-latents, $\hat{x}$ is the compressed image. AE and AD represent arithmetic encoding and decoding, respectively, and Q represents scalar rounding quantization. Dotted lines from component A to component B indicate that the outputs of A parameterize B. | 2202.00198_3.jpg |
arxivcap_1806.08737_7 | Title: Constraining the multiplicity statistics of the coolest brown dwarfs:
binary fraction continues to decrease with spectral type | Caption: Calculation of the probabilities of observing specific binary systems for given model parameters and achieved sensitivity limits. Top: Detection probability for our observed sample (same as Figure <ref>). Middle: Joint model probability density of companion distribution assuming populations described by $\rho_0$ = 4 AU, $\sigma$ = 0.5 and $\gamma$ = 3. The separation distribution was truncated at 1.5 AU and 1000 AU. Bottom: Same as middle panel scaled by the detection probability from the top panel. This provides the companion distribution expected to be observed for the achieved limits and model parameters considered. The probability of observing a system with separation $\rho$ and mass ratio $q$ is given by the volume under the final density function for the region delimited by the black lines which encompasses the observed $\rho$ and $q$ values. | 1806.08737_7.jpg |
arxivcap_2305.05391_5 | Title: Privacy-preserving Adversarial Facial Features | Caption: Transferability of AdvFace on defending against reconstruction attacks. Ours(Trans), Ours(Res), and Ours(Unet) represent that we adopt the transpose, resnet, and unet network as the shadow model, respectively. TransRec, ResRec, and URec represent that the transpose, resnet, and unet network is employed as the reconstruction network, respectively. | 2305.05391_5.jpg |
arxivcap_1912.05016_5 | Title: Robust Feature-Based Point Registration Using Directional Mixture Model | Caption: Example from the KITTI dataset <cit.>. Figures show two urban scenes pointcloud data and associated surface normals on a unit sphere. As seen, as the vehicle moves, the surface normals are also moving on the unit sphere which indicates the rotation between consecutive frames. | 1912.05016_5.jpg |
arxivcap_2303.14753_2 | Title: Does "Deep Learning on a Data Diet" reproduce? Overall yes, but GraNd at Initialization does not | Caption: Reproduction of Figure 1 (second row) from <cit.>. In both reproductions, GraNd at initialization performs as well as the input norm. However, it does not perform better than random pruning. Importantly, it also fails to reproduce the results from <cit.>. However, GraNd at epoch 20 (respectively at epoch 1 for `hlb') performs similar to EL2N and like GraNd at initialization in <cit.>. | 2303.14753_2.jpg |
arxivcap_1702.06441_3 | Title: Comprehensive Data Set for Automatic Single Camera Visual Speed Measurement | Caption: Possible variants of occlusion. a) the vehicles are not occluded at all, b) the closer vehicle is occluding the frontal part of the farther vehicle c) the farther vehicle is fully covered, d) the farther vehicle's rear part is covered. The graphs below represent LIDAR responses with different levels for empty road, fast lane (top dashed line), and slower lane (bottom dashed lane). See text for description of how all these situations are handled. Best viewed on screen. | 1702.06441_3.jpg |
arxivcap_1401.8053_3 | Title: Hallucinating optimal high-dimensional subspaces | Caption: (a) Illuminations 1-7 from the Cambridge face motion database. (b) Five different individuals in the illumination setting number 6. In spite of the same spatial arrangement of light sources, their effect on the appearance of faces changes significantly due to variations in people's heights and the ad lib chosen position relative to the camera. | 1401.8053_3.jpg |
arxivcap_2212.13254_8 | Title: Modeling high-entropy transition-metal alloys with alchemical compression | Caption: Trajectories of the potential energy for the 40 replicas used in one of the REMD simulations of a 864-atoms box of the . The collection of trajectory segments corresponding to the extremal temperatures $T=300$ K and $T=1253$ K are highlighted with thicker, black lines. The logarithmic time scale refers to the MD integration time, but should not be interpreted as physical time given the presence of MC steps and replica exchange moves. | 2212.13254_8.jpg |
arxivcap_2202.13821_2 | Title: Three-dimensional discontinuous Galerkin based high-order gas-kinetic scheme and GPU implementation | Caption: Taylor-Green vortex: iso-surface of the second invariant of velocity gradient tensor at $t = 2.5$, $5$, $7.5$ and $10$ colored by velocity magnitude. | 2202.13821_2.jpg |
arxivcap_2108.09603_5 | Title: Tensor Pooling Driven Instance Segmentation Framework for Baggage Threat Recognition | Caption: Detection performance of the proposed system in terms of mAP (A, B, C), and computational time in terms of seconds (D, E, F) obtained for GDXray <cit.>, SIXray <cit.>, and OPIXray <cit.> datasets, respectively. | 2108.09603_5.jpg |
arxivcap_2305.05391_6 | Title: Privacy-preserving Adversarial Facial Features | Caption: The pipeline of AdvFace in the deployed face recognition system. In the client, the original image is converted to a facial feature which will be uploaded to the server. In the server, the facial feature is reconstructed to a shadow image by the shadow model built by the server, and the shadow feature is extracted and further converted to the reconstructed image. Finally, the generation of the adversarial feature depends on the loss of the shadow image and the reconstructed image. The server stores adversarial features rather than original facial features uploaded from the client against reconstruction attacks. | 2305.05391_6.jpg |
arxivcap_2202.00198_4 | Title: Recognition-Aware Learned Image Compression | Caption: Comparison between the state of the art traditional codec, BPG, and our joint model, in terms of bitrate, accuracy, and PSNR. By traversing the $\lambda$, $\beta$ space we attempt to find equivalent or proximal bitrates to those achieved by BPG. A demonstration of distortion control using parameters $\lambda$, $\beta$. Models with $\beta=1.0$ are trained with initial learning rate 5e-5 and dropout of 0.2; all others are trained using an initial learning rate of 1e-4 with no dropout. A demonstration of accuracy control using parameters $\lambda$, $\beta$. Models are trained using an initial learning rate of 1e-4 with no dropout. | 2202.00198_4.jpg |
arxivcap_2012.05294_1 | Title: A modified Susceptible-Infected-Recovered model for observed under-reported incidence data | Caption: {Histograms of the marginal posterior distribution of the transmission rate $\beta$ (left), fraction of reported cases $p$ (middle) and the negative binomial shape parameter $r$ (right) for each country.} The $x$-axis corresponds to the estimated values, and the $y$-axis is the bin's relative frequency. | 2012.05294_1.jpg |
arxivcap_1401.8053_4 | Title: Hallucinating optimal high-dimensional subspaces | Caption: Examples of 10 roughly angularly equidistant views (out of 73 available) of two objects from the "Object View Collection" subset of the Amsterdam Library of Object Images <cit.>. | 1401.8053_4.jpg |
arxivcap_1912.05127_1 | Title: A Closer Look at Disentangling in $\beta$-VAE | Caption: $\beta$-dependence of various quantities at the optimal parameter configuration of $\beta$-VAE. (A) ELBO as a function of $\beta$. (B) MIE/TIE as a function of $\beta$. (C) Reconstruction objective as a function of $\beta$. (D) Conditional Independence Loss as a function of $\beta$. In these plots, we averaged the plotted quantities over the data distribution. | 1912.05127_1.jpg |
arxivcap_1702.06441_5 | Title: Comprehensive Data Set for Automatic Single Camera Visual Speed Measurement | Caption: Cumulative histograms of distribution of errors. The dashed vertical line represents 3 km/h (or 3 %) threshold. See text for details. (Line markers represent every 1500$^\mathrm{th}$ data sample.) | 1702.06441_5.jpg |
arxivcap_2202.01497_1 | Title: End-to-End Latency Analysis and Optimal Block Size of Proof-of-Work Blockchain Applications | Caption: PoW-based blockchain operation: (1) miners receive transactions from users and other miners, (2) candidate blocks are mined running consensus, (3) mined blocks are propagated, (4) miners append candidate blocks to their ledger copy (forks may occur, leading to inconsistencies). | 2202.01497_1.jpg |
arxivcap_1806.08737_8 | Title: Constraining the multiplicity statistics of the coolest brown dwarfs:
binary fraction continues to decrease with spectral type | Caption: Posterior probability distribution of the binary frequency $f$ for our observed sample of 12 targets obtained with the MCMC sampler described in Section <ref>. The null detection in the program allows us to place an upper limit of $f<10.7$% (1-$\sigma$) on the binary fraction of our survey at separations $>$2 AU. | 1806.08737_8.jpg |
arxivcap_2202.00158_4 | Title: Learning-Based Framework for Camera Calibration with Distortion Correction and High Precision Feature Detection | Caption: Left: Examples of synthetic images. Right: Examples of images under different degrees of distortion($k_0 = 1, -0.5\leq k_1 \leq -0.2, k_2 = -0.02$) (top) along with the corrected images (bottom) acquired by our method. | 2202.00158_4.jpg |
arxivcap_2305.05347_1 | Title: On the onset delays of solar energetic electrons and protons: Evidence for a common accelerator | Caption: The left and middle panels are analytical estimates of the onset delay, for a SEP point source, as a function of magnetic connectivity for protons and electrons, respectively. The right panel shows the relationship between the electron and proton onset delays which is independent of $\lambda_{\perp}$. Observations are taken from <cit.>, while analytical estimates are shown for different assumptions of $\lambda_{\perp}$. | 2305.05347_1.jpg |
arxivcap_1806.08737_9 | Title: Constraining the multiplicity statistics of the coolest brown dwarfs:
binary fraction continues to decrease with spectral type | Caption: Posterior probability distribution of the binary frequency of T8$-$Y0 brown dwarfs on the separation range 1.5$-$1000 AU for the extended sample of 23 objects, combining our observed HST program and the $\geqslant$T8 sources from the surveys in <cit.> and <cit.>. The detection of one companion in the additional subset provides well-defined 1-$\sigma$ and 2-$\sigma$ confidence intervals around a most likely value for the observed binary fraction, $f = 5.2\%$. | 1806.08737_9.jpg |
arxivcap_1802.01192_5 | Title: MOCCA-SURVEY Database I: Assessing GW kick retention fractions for BH-BH mergers in globular clusters | Caption: Time dependent GW kick retention fractions. Top: MS0, middle: MS1, bottom: MS2. The overall trend is increasing with time, and at the late stages of cluster evolution mergers are almost always retained. Primordial mergers are most likely to be kicked out in MS1 and least likely in MS0. | 1802.01192_5.jpg |
arxivcap_2303.14753_3 | Title: Does "Deep Learning on a Data Diet" reproduce? Overall yes, but GraNd at Initialization does not | Caption: Rank Correlations of the Scores. Cf. Figure 12 in the appendix of <cit.>. In both reproductions, GraNd at initialization and input norm are positively correlated, while GraNd and EL2N at later epochs are strongly correlated with each other and the Forget Score (at epoch 200). | 2303.14753_3.jpg |
arxivcap_1401.8053_5 | Title: Hallucinating optimal high-dimensional subspaces | Caption: Different scales used as low resolution matching input for (a) face and (b) object data. Square face images with the widths of 5 to 25 pixels at 5 pixel increments, were considered. Images from the Amsterdam Library of Object Images were sub-sampled to 5% to 25% of the original size, at 5% increments. | 1401.8053_5.jpg |
arxivcap_2202.01497_4 | Title: End-to-End Latency Analysis and Optimal Block Size of Proof-of-Work Blockchain Applications | Caption: Blockchain queuing delay for different $\tau$, $\mu$, $\lambda$, and $b$ parameters: (a) without forks ($M=1$), (b) with forks ($M=10$). | 2202.01497_4.jpg |
arxivcap_2012.05294_2 | Title: A modified Susceptible-Infected-Recovered model for observed under-reported incidence data | Caption: {Daily COVID-19 incidence and fraction of reported cases for the United States of America from January 20, 2020 to May 18, 2021.} | 2012.05294_2.jpg |
arxivcap_2212.13254_9 | Title: Modeling high-entropy transition-metal alloys with alchemical compression | Caption: Pair correlation functions computed on a the $T=300$ K (full) and $T=1253$ K (dashed lines) replicas of a box. Black lines correspond to the unresolved pair correlation, while red (Cr-Cr) and blue (Y-Y) lines provide representative examples of pair correlations resolved by species. The vertical dotted lines indicate the regions used in the definition of the pair ordering. | 2212.13254_9.jpg |
arxivcap_1906.05401_2 | Title: The role of metallic leads and electronic degeneracies in thermoelectric power generation in quantum dots | Caption: (a) Conductance at 3.1K and 32K as a function of $\alpha \lvert e\rvert V_\mathrm{g}$ and the corresponding fits. (b) Conductance as a function of $T$ and $\alpha \lvert e\rvert V_\mathrm{g}$. The dashed line indicates the calculated position of the conductance peaks. (c) Thermovoltage at 3.1K and 32K as a function of $\alpha \lvert e\rvert V_\mathrm{g}$ and the corresponding fit. (d) Thermovoltage as a function of $\alpha \lvert e\rvert V_\mathrm{g}$ and $T$. (e) Schematic of the electrical circuit driven by the thermocurrent $L_{12} \Delta T$, the red dotted line marks the device. (f) Corresponding thermal circuit diagram for our system driven by the heat current $L_{21} \Delta V$. The temperature difference for all calculations: $\Delta T = 0.52$K. | 1906.05401_2.jpg |
arxivcap_2202.00158_5 | Title: Learning-Based Framework for Camera Calibration with Distortion Correction and High Precision Feature Detection | Caption: Our framework includes three main parts. First, radial distortion is removed by correction model inferred from distorted images by our network (green part). The chessboard corner detection part consists of learning heatmap and sub-pixel refinement with outlier rejection and collineation post-processing (blue part). After precise sub-pixel corner coordinates are obtained, stable parameter estimation owning to a simplified RANSAC procedure is performed to achieve accurate camera parameters (gray part). | 2202.00158_5.jpg |
arxivcap_1806.08737_10 | Title: Constraining the multiplicity statistics of the coolest brown dwarfs:
binary fraction continues to decrease with spectral type | Caption: Same as Figure <ref> for the sample of 24 T5$-$T7.5 brown dwarfs compiled from the surveys in <cit.> and <cit.>. The obtained distribution is very similar to that obtained for the sample of late-T objects for the same separation range (1.5$-$1000 AU). | 1806.08737_10.jpg |
arxivcap_2305.05347_2 | Title: On the onset delays of solar energetic electrons and protons: Evidence for a common accelerator | Caption: The modelled peak delay as a function of the onset delay for protons (left panel) and electrons (right panel). The measurements are again taken from <cit.>. {Note that, when no model values are shown, the model intensity did not cross the background level at that longitude.} | 2305.05347_2.jpg |
arxivcap_1912.05127_2 | Title: A Closer Look at Disentangling in $\beta$-VAE | Caption: Values for error terms across 100 random initializations of the network. Solid line represents the average. Dashed lines around the solid line represent the minimum and maximum values, and vertical dashed line represent the extremum. (A) ELBO as a function of $\beta$. (B) TIE as a function of $\beta$. (C) Reconstruction Objective as a function of $\beta$. (D) Conditional Independence Loss as a function of $\beta$. | 1912.05127_2.jpg |
arxivcap_1806.08737_11 | Title: Constraining the multiplicity statistics of the coolest brown dwarfs:
binary fraction continues to decrease with spectral type | Caption: Detection probability map for the combined samples of mid and late-T brown dwarfs, compiled from our observed program and the surveys from <cit.> and <cit.>. Contours denote the 0%, 50% and 100% completeness regions. The red stars show the positions of the two binaries discovered by <cit.>, using the component masses calculated in this work. The yellow star corresponds to the known unresolved W0146$+$4234AB system, located in the 0$-$10% detection probability region of the full sample. | 1806.08737_11.jpg |
arxivcap_1702.06264_5 | Title: Weighted Motion Averaging for the Registration of Multi-View Range Scans | Caption: The objective function value of the registration results for the competed approaches in each MC trial, the sub-graphs (e) and (f) are the amplified results of (c) and (d), respectively | 1702.06264_5.jpg |
arxivcap_2012.05294_3 | Title: A modified Susceptible-Infected-Recovered model for observed under-reported incidence data | Caption: {Daily COVID-19 incidence and fraction of reported cases for Brazil from February 26, 2020 to May 18, 2021.} | 2012.05294_3.jpg |
arxivcap_2206.01855_6 | Title: A possible surviving companion of the SN Ia in the Galactic SNR G272.2-3.2 | Caption: The distributions in $\mu^{*}_{l}$, $\mu_{b}$ (two upper panels) and $v_{r}$ (lower panel) of the M dwarf stars with about solar metallicity and at distances 1 kpc $\leq d \leq$ 2 kpc, in the direction of G272.2-3.2, from the Besançon model of the Galaxy (https://model.obs-besancon.fr; see text for details). | 2206.01855_6.jpg |
arxivcap_2202.01539_1 | Title: Electronic energies from coupled fermionic 'Zombie' states imaginary time evolution | Caption: Imaginary time propagation starting with the 6-electron restricted Hartree Fock Slater determinant using a complete Slater determinant basis (dashed, red line) and a complete random basis (solid, orange line). Both results tend to neutral ground sate energy of Li$_2$ found by diagonalizing the Slater determinant Hamiltonian, shown as a grey dotted line. | 2202.01539_1.jpg |
arxivcap_2202.00231_2 | Title: Are strongly confined colloids good models for two dimensional liquids? | Caption: Comparing diffusion and relaxation of 2D and quasi-2D systems: (a) MSD and (b) SISF as a function of $t$ for six typical $\phi$; the wave-vector $k$ is $k_0=5.0$, close to the the peak of the structure factor $S(k)$; (c) $D_{\rm L}/D_0$ and $\tau_\alpha$ as a function of $\phi$. The inset in (a) displays the log-scale slopes of the solid lines in the main plot. | 2202.00231_2.jpg |
arxivcap_1906.05401_3 | Title: The role of metallic leads and electronic degeneracies in thermoelectric power generation in quantum dots | Caption: (a) Experimental power factor at 3.1K. (b) Calculated power factor for a degenerate (black line) and non-degenerate (dot-dashed green line) energy level, holding all other parameters constant. All as a function of $\alpha \lvert e\rvert V_\mathrm{g}$. (c) Ratio $r_\mathrm{\kappa}$ as a function of $T$. The ratio at each temperature was extracted from fits to the respective thermovoltage <cit.> Inset: maximum thermovoltage as a function of temperature. (d) Maximum power factor $\mathcal{P}^\mathrm{max}$ as a function of $T$. The blue dots show $\mathcal{P}^\mathrm{max}$ calculated with the temperature drop between the contacts ($\Delta T$) and the orange triangles $\mathcal{P}_\mathrm{eff}^\mathrm{max}$ calculated with the effective temperature drop $r_\mathrm{\kappa} \Delta T$. In all plots a temperature difference between the two contacts of $\Delta T \approx 0.52$K is used and the propagated error from the temperature calibration is given by the shaded area. | 1906.05401_3.jpg |
arxivcap_1806.08737_12 | Title: Constraining the multiplicity statistics of the coolest brown dwarfs:
binary fraction continues to decrease with spectral type | Caption: Posterior probability distribution of the $>$1.5 AU binary frequency of $\geqslant$T5 brown dwarfs for the combined sample of 47 objects compiled from our program and the surveys in <cit.> and <cit.>. | 1806.08737_12.jpg |
arxivcap_2303.14673_1 | Title: Type III superconductivity | Caption: Resistance measurements. a: Four-terminal dc resistance measurements of the 20-nm-thick NbN film at different temperatures. The red and blue curves correspond to BKT fitting and VFT fitting to the experimental data, respectively. The gray points mark experimental data deviating from the fits. The inset depicts a sketch of the four-terminal dc resistance measurements of NbN and NbTiN films. Electrodes V$^+$ and V$^-$ correspond to the two voltage probes for measuring the voltage difference. Electrodes I$^+$ and I$^-$ are the source and drain contacts. The symbols V and I stand the voltmeter and ammeter measuring the voltage difference and current, respectively. b: Four-terminal dc resistance measurements of a 20-nm-thick NbN film at different temperatures. The red and blue curves correspond to the BKT fitting and VFT fitting to the experimental data, respectively. The three data points marked gray show the noticeable deviation from the fits. | 2303.14673_1.jpg |
arxivcap_2305.05347_3 | Title: On the onset delays of solar energetic electrons and protons: Evidence for a common accelerator | Caption: {The left panel shows the assumed slab turbulence spectrum at Earth, as a function of the parallel wavenumber, while the shaded regions indicate where electrons (red shading) and protons (blue shading) in the assumed energy ranges will resonate}. {The right panel shows the} resulting electron and proton parallel and perpendicular mean-free-paths as used in this study. Also shown is the magnetic focusing length. | 2305.05347_3.jpg |
arxivcap_2212.13254_10 | Title: Modeling high-entropy transition-metal alloys with alchemical compression | Caption: A plot of the relative pair probability for all atom pairs and the three regions corresponding to the first, second, and third peaks in the total pair correlation function ( Fig. <ref>). Each plot shows results for simulations of at both 300 K (lower-left corner) and 1253 K (top-right corner), averaged over the trajectories and discarding the first 100 ps (50'000 combined MD/MC steps). | 2212.13254_10.jpg |
arxivcap_2012.05294_4 | Title: A modified Susceptible-Infected-Recovered model for observed under-reported incidence data | Caption: {Daily COVID-19 incidence and fraction of reported cases for Peru from March 07, 2020 to May 18, 2021.} | 2012.05294_4.jpg |
arxivcap_1806.08737_13 | Title: Constraining the multiplicity statistics of the coolest brown dwarfs:
binary fraction continues to decrease with spectral type | Caption: Stellar and substellar binary fractions as a function of the spectral type in the Galactic field showing a clear decline in binary frequency with decreasing primary mass. Filled symbols correspond to total binary fractions, estimated over complete ranges of separations and mass ratios. The $>$1.5 AU (open symbol) and overall (filled symbol) binary frequencies determined in this work for the T5$-$Y0 population are shown in red. The data point from <cit.> corresponds to the upper limit estimated in that paper for a lognormal distribution in separation and a power law in mass ratio, similar to the distributions considered here. Error bars on binary fractions correspond to 1-$\sigma$ uncertainties, except for the value from <cit.> which corresponds to a 2-$\sigma$ confidence level (see text). The ranges of separations and mass ratios considered in each survey are plotted in Figure <ref>. Some points were slightly shifted to make the figure clearer. | 1806.08737_13.jpg |
arxivcap_1401.8053_6 | Title: Hallucinating optimal high-dimensional subspaces | Caption: The increase in class separation $\mu$ (the ordinate; note that the scale is logarithmic) over different training-query illumination conditions (abscissa), achieved by the proposed method in comparison to the naïve subspace re-projection approach. Note that for clarity the training-query illumination pairs were ordered in increasing order of improvement for each plot; thus, the indices of different abscissae do not necessarily correspond. | 1401.8053_6.jpg |
arxivcap_1912.05127_3 | Title: A Closer Look at Disentangling in $\beta$-VAE | Caption: Values for error terms across 300 random initializations of the network for a synthetic dataset, which comprises of a single MNIST digit localized at different locations on a blank canvas. The cartesian coordinate of the digit in a sample from our data, $\x$, is determined by eq. (<ref>), with $A_{ij} = 2\delta_{ij} + 0.73$, $\s \sim \N(\bm{0},\I_k)$, $ \bm{\eta} \sim \N(\bm{0},\I_N)$, $N=k=2$. Dashed lines represent the minimum and maximum values, and solid line represents the average. (A) ELBO as a function of $\beta$. (B) TIE as a function of $\beta$. Its minima over various random initialization follow a non-monotonic trend. (C) Reconstruction objective as a function of $\beta$. (D) Conditional Independence Loss as a function of $\beta$. (E) Traversal of latent encoding in bottleneck neurons for small and large $\beta$. One neuron is held fixed while the other is modulated to generate reconstructions. Reconstruction of the digits noticeable worsens with higher $\beta$, while units in the bottleneck encode for structured, orthogonal axes of motion. | 1912.05127_3.jpg |
arxivcap_2202.00231_3 | Title: Are strongly confined colloids good models for two dimensional liquids? | Caption: Comparing diffusion and relaxation of A and B particles in 2D and quasi-2D systems: (a) MSD and (b) SISF as a function of $t$ for two values of $\phi$; (c) Enhancement factor (defined in the text) of $D_{\rm L}/D_0$ and of $\tau_\alpha$ as a function of $\phi$. | 2202.00231_3.jpg |
arxivcap_1906.05401_6 | Title: The role of metallic leads and electronic degeneracies in thermoelectric power generation in quantum dots | Caption: Contact calibration for the source (a) and drain (b) contact. The measured resistance as a funciton of temperature at 0 mW heating power is shown by the dashed line while the resistance at 2.3 mW heating power is symbolized by the dots. Both curves are offset by 0.2 $\Omega$ for clarity. We then plot the 0 mW heating power curve (squares) overlay ed with a shifted 2.3 mW curve, giving us the temperature increase due to Joule heating of the microheater. | 1906.05401_6.jpg |
arxivcap_2303.14673_2 | Title: Type III superconductivity | Caption: Magnetization measurements under zero-field-cooled (ZFC) and field-cooled (FC) condictions. a: Magnetization of the 20 nm-thick NbN film. Insert: magnetic susceptibility. The measurements are taken under the external field of 20 Oe.b: Magnetization of the 86 nm-thick NbTiN film. The measurements are taken under the external field of 5 Oe. | 2303.14673_2.jpg |
arxivcap_2202.01539_2 | Title: Electronic energies from coupled fermionic 'Zombie' states imaginary time evolution | Caption: Imaginary time propagation starting with the 6 electron restricted Hartree Fock Slater determinant and using a reduced random basis of 200 Zombie sates. The neutral ground sate energy of Li$_2$ found by diagonalising the Slater determinant Hamiltonian is also shown in grey. | 2202.01539_2.jpg |
arxivcap_2012.05294_5 | Title: A modified Susceptible-Infected-Recovered model for observed under-reported incidence data | Caption: {Daily COVID-19 incidence and fraction of reported cases for Mexico from February 28, 2020 to May 18, 2021.} | 2012.05294_5.jpg |
arxivcap_2108.09603_6 | Title: Tensor Pooling Driven Instance Segmentation Framework for Baggage Threat Recognition | Caption: Qualitative evaluations of the proposed framework on GDXray <cit.> dataset. Please zoom-in for best visualization. | 2108.09603_6.jpg |
arxivcap_2305.05347_5 | Title: On the onset delays of solar energetic electrons and protons: Evidence for a common accelerator | Caption: The histograms show the observed distributions of the connection angle (left panels) and full-width half maximum (FWHM; right panels) of 14 - 24 MeV protons (top panels; in blue) and 0.7 - 4 MeV electrons (bottom panels; in red) from Fig. 24 of <cit.>. Observational averages are indicated by the filled circles, while the open green symbols show the results from the best-fit numerical model. | 2305.05347_5.jpg |
arxivcap_2212.13254_11 | Title: Modeling high-entropy transition-metal alloys with alchemical compression | Caption: (a) Element similarity matrix based on the RPP distance (<ref>) for the nearest-neighbor shell, in the simulation at $T=1253$ K. (b) The element similarity map (color-coded based on the group of the various transition metals) is built by applying metric multi-dimensional scaling to the distance matrix, and provides a visual aid to recognize groups of elements that have similar affinity patterns to the other $d$-block metals. | 2212.13254_11.jpg |
arxivcap_1111.0776_2 | Title: Light-cone-like spreading of correlations in a quantum many-body system | Caption: {Time evolution of the two-point parity correlations.} After the quench, a positive correlation signal propagates with increasing time to larger distances. The experimental values for a quench from $U/J=40$ to $U/J=9.0$ (circles) are in good agreement with the corresponding numerical simulation for an infinite, homogeneous system at zero temperature (continuous line). Our analytical model (dashed line) also qualitatively reproduces the observed dynamics. Inset: Experimental data displayed as a colormap, revealing the propagation of the correlation signal with a well defined velocity. The experimental values result from the average over the central $N$ sites of more than 1000 chains, where $N$ equals $80\,\%$ of the length of each chain. Error bars represent the standard deviation. | 1111.0776_2.jpg |
arxivcap_2202.00231_4 | Title: Are strongly confined colloids good models for two dimensional liquids? | Caption: Unifying diffusion and relaxation dynamics in 2D and quasi-2D systems: $D_{\rm L}/D_0$ and $\tau_\alpha$ as a function of (a) $g(r_{\rm 3rd})$ and (b) $\gamma_{\rm min}$; (c) MSD and SISF as a function of $t$ for $g(r=r_{\rm 3rd})\approx 1.7$, 3.0 and 3.5, from normal to glassy (the corresponding area fractions are given in the legend). | 2202.00231_4.jpg |
arxivcap_1906.05275_1 | Title: Keeping Notes: Conditional Natural Language Generation with a Scratchpad Mechanism | Caption: The Scratchpad Mechanism first computes the update probability ($\alpha^i_t$) for each encoder state according to Eq. <ref>, then computes a global update $\ut^i$ according to Eq. <ref>, and finally updates the encoder states according to Eq. <ref>. | 1906.05275_1.jpg |
arxivcap_1802.01192_6 | Title: MOCCA-SURVEY Database I: Assessing GW kick retention fractions for BH-BH mergers in globular clusters | Caption: Time dependent GW kick retention fractions for models with IMBH or BH subsystem at large time scales. Top: MS0, middle: MS1, bottom: MS2. GCs which form IMBH have highest GW kick retention fractions at all time scales, which is due to low $q$ and partly due to low cluster potential resulting in higher escape velocities. A subsystem has a decreasing impact on GW kick retention fractions. | 1802.01192_6.jpg |
arxivcap_2303.14707_1 | Title: Clean-NeRF: Reformulating NeRF to account for View-Dependent Observations | Caption: {Foggy} vs {Clear} NeRF. Our gets rid of reconstruction errors manifested as foggy "floaters" in the density volume without additional input or significant computational overhead. Below are density profiles along a given ray before and after our geometry correction procedure, where we discard density peaks corresponding to floaters. | 2303.14707_1.jpg |
arxivcap_2012.05294_6 | Title: A modified Susceptible-Infected-Recovered model for observed under-reported incidence data | Caption: {Daily COVID-19 incidence and fraction of reported cases for Argentina from March 03, 2020 to May 18, 2021.} | 2012.05294_6.jpg |
arxivcap_1702.06506_1 | Title: PixelNet: Representation of the pixels, by the pixels, and for the pixels | Caption: PixelNet: We input an image to a convolutional neural network, and extract hypercolumn descriptor for a sampled pixel from multiple convolutional layers. The hypercolumn descriptor is then fed to a multi-layer perceptron (MLP) for the non-linear optimization, and the last layer of MLP outputs the required response for the task. See text for more details about the use of network at training/test time. | 1702.06506_1.jpg |
arxivcap_1912.05163_3 | Title: TANet: Robust 3D Object Detection from Point Clouds with Triple Attention | Caption: The architecture of Coarse-to-Fine Regression. The Pyramid Sampling indicates a series of down-sampling and up-sampling operations, which can be achieved via the pooling and the transposition convolution. | 1912.05163_3.jpg |
arxivcap_1806.08737_14 | Title: Constraining the multiplicity statistics of the coolest brown dwarfs:
binary fraction continues to decrease with spectral type | Caption: WFC3/IR F127M (left) and F139M (right) images of W2220$-$3628. The science target is in the yellow circle. The identified candidate companion is encircled in red. While most objects in the field of view have roughly similar fluxes in the two bandpasses, the Y dwarf and the selected candidate both exhibit a strong magnitude drop in the water-band F139M filter. | 1806.08737_14.jpg |
arxivcap_1906.05275_2 | Title: Keeping Notes: Conditional Natural Language Generation with a Scratchpad Mechanism | Caption: Attention over the source sequence visualized at each decoder timestep with and without the scratchpad mechanism. Darker cells mean higher values. "@@" at the end of a bpe token denotes it should be concatenated with the following token(s) to make a word. With Scratchpad, we see sharper attention earlier in the sentence that better aligns with word-level translations (e.g. 'hand' is properly attended to with scratchpad, but not with non-scratchpad). Additionally, some words that are never properly translated (e.g. wahrscheinlich - 'probably') by the non-scratchpad model are not heavily attended to, whereas with the scratchpad mechanism, they are. | 1906.05275_2.jpg |
arxivcap_2202.00231_7 | Title: Are strongly confined colloids good models for two dimensional liquids? | Caption: Results of test simulations of 3D monodisperse colloidal suspensions: (a) The normalized short time diffusion coefficient $D_{\rm S}/D_0$ and (b) the long time diffusion coefficient $D_{\rm L}/D_0$ as a function of $\phi$ for LD and FLD and a series of masses. Also included are the results from Ref. <cit.>.
| 2202.00231_7.jpg |
arxivcap_1702.06506_2 | Title: PixelNet: Representation of the pixels, by the pixels, and for the pixels | Caption: Given a fixed number of (5) images per SGD batch, we analyse convergence properties using all pixels vs. randomly sampling $4\%$ , or $2,000$ pixels for semantic segmentation and surface normal estimation. This experiment suggest that sampling does not hurt convergence. | 1702.06506_2.jpg |
arxivcap_2202.01539_4 | Title: Electronic energies from coupled fermionic 'Zombie' states imaginary time evolution | Caption: Occupational probability plotted against orbital number for five spin orbitals. The first four spin orbitals are set exactly as occupied, the rest are set using the Gaussian described in table <ref>. The average probability is shown by a cross and one standard deviation either side is the bar. | 2202.01539_4.jpg |
arxivcap_2305.05347_6 | Title: On the onset delays of solar energetic electrons and protons: Evidence for a common accelerator | Caption: The left panel shows the maximum omni-directional intensity as a function of the magnetic connection angle for best-fit model assuming a source width of $\sigma = 25^{\circ}$ indicated by the black dashed curve. The right panel shows the resulting intensities for different assumptions of the source size (indicated in the legend). | 2305.05347_6.jpg |
arxivcap_2108.09603_7 | Title: Tensor Pooling Driven Instance Segmentation Framework for Baggage Threat Recognition | Caption: Qualitative evaluations of the proposed framework on SIXray <cit.> dataset. Please zoom-in for a best visualization. | 2108.09603_7.jpg |
arxivcap_2012.05294_7 | Title: A modified Susceptible-Infected-Recovered model for observed under-reported incidence data | Caption: {Daily COVID-19 incidence and fraction of reported cases for Chile from March 03, 2020 to May 18, 2021.} | 2012.05294_7.jpg |
arxivcap_1111.0776_3 | Title: Light-cone-like spreading of correlations in a quantum many-body system | Caption: {Propagation velocity.} a, Determination of the propagation velocity for the quenches to $U/J=5.0$ (triangles), 7.0 (squares) and 9.0 (circles). The time of the maximum of the correlation signal is obtained from fits to the traces $C_{d}(t)$. Error bars represent the 68 % confidence interval of these fits. We then extract the propagation velocities from weigthed linear fits restricted to $2 \leq d \leq 6$ (lines). The data for $U/J = 5.0$ and 7.0 have been offset horizontally for clarity. b, Comparison of the experimental velocities (circles) to the ones obtained from numerical simulations for an infinite, homogeneous system at zero temperature (shaded area). The shaded area and the vertical error bars denote the 68 % confidence interval of the fit. The horizontal error bars represent the uncertainty due to the calibration of the lattice depth. The black line corresponds to the bound $v_{\text{max}}$ predicted by our effective model (the fading indicates the break down of this model). The arrows mark the maximum velocity expected in the non-interacting case (left) and the asymptotic value derived from our model when (right). | 1111.0776_3.jpg |
arxivcap_2206.01855_8 | Title: A possible surviving companion of the SN Ia in the Galactic SNR G272.2-3.2 | Caption: 3D orbit of star M5-G272 along 10$^{9}$ yr (forward in time) and its projections on the XY, XZ and YZ planes of the referential system of the Galaxy. | 2206.01855_8.jpg |
arxivcap_1912.05163_4 | Title: TANet: Robust 3D Object Detection from Point Clouds with Triple Attention | Caption: Qualitative detection results on the KITTI test set. The first and the third row show the 3D bounding boxes projected on the 2D images. The second and the fourth row depict the predicted 3D bounding boxes and corresponding orientations for 3D objects on LiDAR. Cars, Pedestrians and Cyclists are visualized with blue, green and red bounding boxes, respectively. | 1912.05163_4.jpg |
arxivcap_1806.08737_16 | Title: Constraining the multiplicity statistics of the coolest brown dwarfs:
binary fraction continues to decrease with spectral type | Caption: Detection limits reached around our target W0148$-$7202 in terms of minimum detectable masses (left) and mass ratios (right). The magnitude limits for the target (Figure <ref>) were converted to masses using the AMES-COND evolutionary models from <cit.> at discrete ages of 1, 5 and 10 Gyr. The primary mass was calculated at each age considered when converting the mass limits into mass ratios. The bottom panels show the relative scatter in the masses and mass ratios between adopted ages of 1, 5 and 10 Gyr. The mass limit curves have a mean relative scatter of 0.49. The same curves in mass ratio space have a mean relative scatter of 0.11, therefore significantly reducing the uncertainty introduced by adopting a discrete age of 5 Gyr. | 1806.08737_16.jpg |
arxivcap_2212.13254_12 | Title: Modeling high-entropy transition-metal alloys with alchemical compression | Caption: a. Cowley's short-range (SRO) parameters for the first shell in CoCrFeMnNi HEA, shown for the 10 replicas between 300 and 1253 K, averaged over the last 1000 steps and two independent runs. At low temperatures, a tendency of Fe-Mn segregation can be seen. In contrast, Cr is very well mixed. There are two phase transformations around 400 K and 900 K. The y-axis is adjusted to the example shown in Fig. <ref> to facilitate comparison. b,c. snapshot from MC/MD simulations at $T=300\,\si{\kelvin}$ and at $T=720\,\si{\kelvin}$, respectively. In the 300 K snapshot, two planes of Ni can be seen, while in the higher temperature snapshot, Cr order is evident (see the ). | 2212.13254_12.jpg |
arxivcap_2303.14707_2 | Title: Clean-NeRF: Reformulating NeRF to account for View-Dependent Observations | Caption: Appearance Decomposition. (a, d) For object surfaces with strong view-dependent effects, vanilla NeRF often gets disrupted, resulting in poor reconstruction quality. We propose decomposing appearance into (b) view-independent and (c) view-dependent components during NeRF training and incorporating geometry-related priors to improve reconstruction quality. Our approach shows the ability to (e) render the view-independent component of the scene and (f) more accurate recovery of the scene. | 2303.14707_2.jpg |
arxivcap_1702.06506_3 | Title: PixelNet: Representation of the pixels, by the pixels, and for the pixels | Caption: Results on BSDS <cit.>. While our curve is mostly overlapping with HED, our detector focuses on more high-level semantic edges. See qualitative results in Fig.<ref>. | 1702.06506_3.jpg |
arxivcap_2005.06102_7 | Title: Semantic prefetching using forecast slices | Caption: Ratio of injected operations out of the overall, compared to speedup. Showing applications with speedup $\geq$ 15%. | 2005.06102_7.jpg |
arxivcap_1906.05326_2 | Title: Magnetism and superconductivity in Fe$_{1+y}$Te$_{1-x}$Se$_x$ | Caption: Left: Illustration of tetrahedral coordination of Fe by Te/Se. Right: Crystal-field scheme of the relative energies of the Fe $3d$ states in the tetrahedral environment, including the effect of possible distortions, such as orthorhombic strain $(b-a)$ or distortion of the bond angle $\alpha$ from the ideal $109.5^\circ$. | 1906.05326_2.jpg |
arxivcap_2202.00231_8 | Title: Are strongly confined colloids good models for two dimensional liquids? | Caption: Test whether various structural quantities can be used to establish a mapping between the quasi-2D and the 2D systems. $D_{\rm L}/D_0$ as a function of (a) effective area packing fraction $\Phi$, (b) $g(r_{\rm 1st}$), (c) $g(r_{\rm 2nd}$), and (d) $g_{\rm BB}(r_{\rm peak})$. | 2202.00231_8.jpg |
arxivcap_2005.06102_13 | Title: Semantic prefetching using forecast slices | Caption: Dynamic flow of graph500 (Figure <ref>) unrolled to show a possible lookahead prefetch based on the forecast slice. Changing the stride at the beginning of the slice can predict accesses on far iterations. Furthermore, iterations with similar contexts (such as branch history) will invoke their future counterparts. | 2005.06102_13.jpg |
arxivcap_1806.08737_17 | Title: Constraining the multiplicity statistics of the coolest brown dwarfs:
binary fraction continues to decrease with spectral type | Caption: Top: Minimum masses detectable around our 12 targets at the 5-$\sigma$ level as a function of angular (left) and physical (right) projected separation. Magnitude contrasts were converted to masses using the AMES-COND evolutionary models from <cit.> at an adopted age of 5 Gyr. The solid black lines show the mean detection levels for the sample and the 1-$\sigma$ standard deviations around the mean (dotted lines). The shaded region on the left panel represents the pixel scale of the WFC3/IR camera. On the right panel, the mean was calculated starting at the smallest physical separation resolved for all targets. Bottom: Same as top panel for minimum detectable mass ratios. Mass limits were converted to mass ratios using the masses calculated in this work and listed in Table <ref>. | 1806.08737_17.jpg |
arxivcap_1401.8053_7 | Title: Hallucinating optimal high-dimensional subspaces | Caption: Typical similarity matrices resulting from the naïve (left) and the proposed (right) matching approaches. Our method produces a dramatic improvement in class separation as witnessed by the increased dominance of the diagonal elements in the aforementioned matrix. | 1401.8053_7.jpg |
arxivcap_2012.05294_8 | Title: A modified Susceptible-Infected-Recovered model for observed under-reported incidence data | Caption: {Daily COVID-19 incidence and fraction of reported cases for Colombia from March 06, 2020 to May 18, 2021.} | 2012.05294_8.jpg |
arxivcap_1111.0776_4 | Title: Light-cone-like spreading of correlations in a quantum many-body system | Caption: Time evolution of the two-point parity correlations. Left panel: quench to $U/J = 5.0(2)$. Right panel: quench to $U/J = 7.0(3)$. The circles indicate the correlations measured experimentally and the line is derived from the numerical simulations for an infinite, homogeneous system at zero temperature. The experimental and numerical values were obtained in the same way as described in the legend of Fig. <ref> and in the Methods Summary section. | 1111.0776_4.jpg |
arxivcap_1802.01192_7 | Title: MOCCA-SURVEY Database I: Assessing GW kick retention fractions for BH-BH mergers in globular clusters | Caption: Relation between $\chi_{\text{eff}}$ and $a_{\text{fin}}$, top MS0, bottom MS1 (for MS2 the plot would not be interesting, since in MS2 every BH (except for post-merger BHs) has the same spin). For primordial mergers there is a clear quasi-linear relationship between $\chi_{\text{eff}}$ and $a_{\text{fin}}$. Points deviating from this relationship correspond to $q\lesssim0.8$. For dynamical mergers constraints introduced by MS1 model lead to a very interesting "cross" structure, related to the non-continuous dependence on metallicity. On the bottom picture vertical lines are at $-0.25$, $-0.13$, $0.13$ and $0.25$, horizontal lines at $0.13$ and $0.25$, they indicate areas of $\left|\chi_{\text{eff}}\right|=a_{\text{low}}$, and $a_{\text{fin}}=a_{\text{low}}$ for Z ranges D ($0.13$) and A and C ($0.25$). | 1802.01192_7.jpg |
arxivcap_2212.13254_13 | Title: Modeling high-entropy transition-metal alloys with alchemical compression | Caption: a. Cowley's short-range (SRO) parameters for the first shell in CoCrFeMoNi HEA, shown for the 10 replicas between 300 and 1253 K, averaged over the last 1000 steps and two independent runs. Good mixing of atomic species can be assumed due to the small values of SRO parameters. The y-axis is adjusted to the example shown in Fig. <ref> to facilitate comparison. b,c. snapshot from MC/MD simulations at $T=300\,\si{\kelvin}$ and at $T=1253\,\si{\kelvin}$, respectively. In the 300 K snapshot, two planes of Ni can be seen. | 2212.13254_13.jpg |
arxivcap_2202.01539_5 | Title: Electronic energies from coupled fermionic 'Zombie' states imaginary time evolution | Caption: Imaginary time propagation for two biased bases each made up of 63 biased Zombie state basis functions and then a Zombie state set either to the 6 electron restricted Hartree Fock determinant (orange bottom line) or the 7 electron open-shell restricted Hartree Fock determinant (red top line). The basis with six electron determinant evolves to the Li$_2$ neutral ground state and the basis with seven electrons evolves to the higher $Li_2^-$ anion ground state. The corresponding energies, obtained by diagonalising the complete Slater determinant basis, are also shown as horizontal lines. The details of how each electron is biased are summarised in Table <ref>. | 2202.01539_5.jpg |
arxivcap_2005.06102_16 | Title: Semantic prefetching using forecast slices | Caption: Breakdown of slices in Figure <ref> according to their load-chain depth and number of arithmetic operations, measured over SPEC 2006/2017. The front line marks the rough range covered by existing prefetchers. | 2005.06102_16.jpg |
arxivcap_2303.14707_6 | Title: Clean-NeRF: Reformulating NeRF to account for View-Dependent Observations | Caption: Qualitative Comparison between and baselines. recovers intricate object details while removing annoying "floaters" even when observations are sparse. | 2303.14707_6.jpg |
arxivcap_2108.09603_9 | Title: Tensor Pooling Driven Instance Segmentation Framework for Baggage Threat Recognition | Caption: Failure cases from GDXray <cit.>, SIXray <cit.>, and OPIXray <cit.> dataset. | 2108.09603_9.jpg |
arxivcap_1906.05326_3 | Title: Magnetism and superconductivity in Fe$_{1+y}$Te$_{1-x}$Se$_x$ | Caption: Phase diagram of with $y \ge0$ as a function of $x$ and $T$ from <cit.>. The nominal Fe content, $y$, is zero unless specified otherwise. Blue circles: superconducting transition temperature, $T_c$; red circles: transition to long-range spin-density-wave order, $T_{\rm SDW}$; orange circles: onset of short-range spin-glass order, $T_{\rm SG}$. 2010 The Physical Society of Japan. | 1906.05326_3.jpg |
arxivcap_1401.8053_8 | Title: Hallucinating optimal high-dimensional subspaces | Caption: Mean class separation increase achieved, as a function of size of the low-scale images. Shown is the ratio of class separation when subspaces are matched using the proposed method and the naïve re-projection method described in Section <ref>. The rate of improvement decay is incrementally exponential, reaching 1 (no improvement) when $d_l = d_h$. | 1401.8053_8.jpg |
arxivcap_1912.05163_5 | Title: TANet: Robust 3D Object Detection from Point Clouds with Triple Attention | Caption: The full pipeline of TANet. First, we equally divide the point clouds into a voxel grid consisting of a set of voxels. Then, the stacked triple attention separately process each voxel to obtain a more discriminative representation. Subsequently, a compact feature representation for each voxel is extracted by aggregating the points inside it in a max-pooling manner. We arrange the voxel feature according to its original spatial position in the grid and thus lead to a feature representation for the voxel grid in the shape of $C' \times H \times W$. Finally, the coarse-to-fine regression is employed to generate the final 3D bounding boxes. | 1912.05163_5.jpg |
arxivcap_2005.06318_1 | Title: A 28-nm Convolutional Neuromorphic Processor Enabling Online Learning with Spike-Based Retinas | Caption: Summary of bottom-up and top-down neuromorphic design approaches. | 2005.06318_1.jpg |
arxivcap_2202.13852_1 | Title: Hyperbolic Graph Neural Networks: A Review of Methods and Applications | Caption: Illustration of three prevalent and isomorphic hyperbolic models: Lorentz model, Kelin model, and Poincaré ball model. | 2202.13852_1.jpg |
arxivcap_1702.06506_4 | Title: PixelNet: Representation of the pixels, by the pixels, and for the pixels | Caption: Segmentation results on PASCAL-Context 59-class. Our approach uses an MLP to integrate information from both lower ($1_2$) and higher (conv-7) layers, which allows us to better capture both global structure (object/scene layout) and fine details (small objects) compared to FCN-8s. | 1702.06506_4.jpg |
arxivcap_2202.00231_9 | Title: Are strongly confined colloids good models for two dimensional liquids? | Caption: Results of test simulations of 2D and quasi-2D systems with LD: $D_{\rm L}/D_0$ and $\tau_\alpha$ as a function of (a) $\phi$, (b) $g(r_{\rm 3rd}$) and (c) $\gamma_{\rm min}$. | 2202.00231_9.jpg |
arxivcap_1806.08737_18 | Title: Constraining the multiplicity statistics of the coolest brown dwarfs:
binary fraction continues to decrease with spectral type | Caption: Mass ratio sensitivity limits for the additional late-T sample. Top: minimum mass ratios detectable around the 7 targets from <cit.> at the 5-$\sigma$ level in the Keck/NIRC2 $H$-band images (grey lines), using the masses calculated in this work and listed in Table <ref>. The shaded region on the left panel represents 4 pixels on the Keck/NIRC2 narrow camera, the radius at which the contrast curves were started. The black solid line shows the mean sensitivity level for the subset and the 1-$\sigma$ standard deviation around the mean (dotted lines). In the right panel, the mean and standard deviation were only calculated at physical separations resolved for all targets. The blue star indicates the position of the secondary companion W0458$+$6434B. The binary companion was masked before computing the contrast curve around the primary W0458$+$6434A. Bottom: same as top panels for the 4 targets from <cit.> in the HST/WFC3 F127M images. The shaded region on the left panel represents the WFC3/IR plate scale. | 1806.08737_18.jpg |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.