arXiv subjects

Zhongyang Zhang

Publications and source records attributed to Zhongyang Zhang.

13 recordsLinked to original sources

arXiv2025

Predicting Quality of Video Gaming Experience Using Global-Scale Telemetry Data and Federated Learning

Frames Per Second (FPS) significantly affects the gaming experience. Providing players with accurate FPS estimates prior to purchase benefits both players and game developers. However, we have a limited understanding of how to predict a game's FPS performance on a specific device. In this paper, we first conduct a comprehensive analysis of a wide range of factors that may affect game FPS on a global-scale dataset to identify the determinants of FPS. This includes player-side and game-side characteristics, as well as country-level socio-economic statistics. Furthermore, recognizing that accurate FPS predictions require extensive user data, which raises privacy concerns, we propose a federated learning-based model to ensure user privacy. Each player and game is assigned a unique learnable knowledge kernel that gradually extracts latent features for improved accuracy. We also introduce a novel training and prediction scheme that allows these kernels to be dynamically plug-and-play, effectively addressing cold start issues. To train this model with minimal bias, we collected a large telemetry dataset from 224 countries and regions, 100,000 users, and 835 games. Our model achieved a mean Wasserstein distance of 0.469 between predicted and ground truth FPS distributions, outperforming all baseline methods.

cs.HC↗

arXiv2024

Labits: Layered Bidirectional Time Surfaces Representation for Event Camera-based Continuous Dense Trajectory Estimation

Event cameras provide a compelling alternative to traditional frame-based sensors, capturing dynamic scenes with high temporal resolution and low latency. Moving objects trigger events with precise timestamps along their trajectory, enabling smooth continuous-time estimation. However, few works have attempted to optimize the information loss during event representation construction, imposing a ceiling on this task. Fully exploiting event cameras requires representations that simultaneously preserve fine-grained temporal information, stable and characteristic 2D visual features, and temporally consistent information density, an unmet challenge in existing representations. We introduce Labits: Layered Bidirectional Time Surfaces, a simple yet elegant representation designed to retain all these features. Additionally, we propose a dedicated module for extracting active pixel local optical flow (APLOF), significantly boosting the performance. Our approach achieves an impressive 49% reduction in trajectory end-point error (TEPE) compared to the previous state-of-the-art on the MultiFlow dataset. The code will be released upon acceptance.

cs.CV↗

arXiv2024

V2CE: Video to Continuous Events Simulator

Dynamic Vision Sensor (DVS)-based solutions have recently garnered significant interest across various computer vision tasks, offering notable benefits in terms of dynamic range, temporal resolution, and inference speed. However, as a relatively nascent vision sensor compared to Active Pixel Sensor (APS) devices such as RGB cameras, DVS suffers from a dearth of ample labeled datasets. Prior efforts to convert APS data into events often grapple with issues such as a considerable domain shift from real events, the absence of quantified validation, and layering problems within the time axis. In this paper, we present a novel method for video-to-events stream conversion from multiple perspectives, considering the specific characteristics of DVS. A series of carefully designed losses helps enhance the quality of generated event voxels significantly. We also propose a novel local dynamic-aware timestamp inference strategy to accurately recover event timestamps from event voxels in a continuous fashion and eliminate the temporal layering problem. Results from rigorous validation through quantified metrics at all stages of the pipeline establish our method unquestionably as the current state-of-the-art (SOTA).

cs.CV↗

arXiv2024

Spike-based Neuromorphic Computing for Next-Generation Computer Vision

Neuromorphic Computing promises orders of magnitude improvement in energy efficiency compared to traditional von Neumann computing paradigm. The goal is to develop an adaptive, fault-tolerant, low-footprint, fast, low-energy intelligent system by learning and emulating brain functionality which can be realized through innovation in different abstraction layers including material, device, circuit, architecture and algorithm. As the energy consumption in complex vision tasks keep increasing exponentially due to larger data set and resource-constrained edge devices become increasingly ubiquitous, spike-based neuromorphic computing approaches can be viable alternative to deep convolutional neural network that is dominating the vision field today. In this book chapter, we introduce neuromorphic computing, outline a few representative examples from different layers of the design stack (devices, circuits and algorithms) and conclude with a few exciting applications and future research directions that seem promising for computer vision in the near future.

cs.NE↗

arXiv2023

"Reading Between the Heat": Co-Teaching Body Thermal Signatures for Non-intrusive Stress Detection

Stress impacts our physical and mental health as well as our social life. A passive and contactless indoor stress monitoring system can unlock numerous important applications such as workplace productivity assessment, smart homes, and personalized mental health monitoring. While the thermal signatures from a user's body captured by a thermal camera can provide important information about the "fight-flight" response of the sympathetic and parasympathetic nervous system, relying solely on thermal imaging for training a stress prediction model often lead to overfitting and consequently a suboptimal performance. This paper addresses this challenge by introducing ThermaStrain, a novel co-teaching framework that achieves high-stress prediction performance by transferring knowledge from the wearable modality to the contactless thermal modality. During training, ThermaStrain incorporates a wearable electrodermal activity (EDA) sensor to generate stress-indicative representations from thermal videos, emulating stress-indicative representations from a wearable EDA sensor. During testing, only thermal sensing is used, and stress-indicative patterns from thermal data and emulated EDA representations are extracted to improve stress assessment. The study collected a comprehensive dataset with thermal video and EDA data under various stress conditions and distances. ThermaStrain achieves an F1 score of 0.8293 in binary stress classification, outperforming the thermal-only baseline approach by over 9%. Extensive evaluations highlight ThermaStrain's effectiveness in recognizing stress-indicative attributes, its adaptability across distances and stress scenarios, real-time executability on edge platforms, its applicability to multi-individual sensing, ability to function on limited visibility and unfamiliar conditions, and the advantages of its co-teaching approach.

cs.HC↗

arXiv2023

Neuromorphic High-Frequency 3D Dancing Pose Estimation in Dynamic Environment

As a beloved sport worldwide, dancing is getting integrated into traditional and virtual reality-based gaming platforms nowadays. It opens up new opportunities in the technology-mediated dancing space. These platforms primarily rely on passive and continuous human pose estimation as an input capture mechanism. Existing solutions are mainly based on RGB or RGB-Depth cameras for dance games. The former suffers in low-lighting conditions due to the motion blur and low sensitivity, while the latter is too power-hungry, has a low frame rate, and has limited working distance. With ultra-low latency, energy efficiency, and wide dynamic range characteristics, the event camera is a promising solution to overcome these shortcomings. We propose YeLan, an event camera-based 3-dimensional high-frequency human pose estimation(HPE) system that survives low-lighting conditions and dynamic backgrounds. We collected the world's first event camera dance dataset and developed a fully customizable motion-to-event physics-aware simulator. YeLan outperforms the baseline models in these challenging conditions and demonstrated robustness against different types of clothing, background motion, viewing angle, occlusion, and lighting fluctuations.

cs.CV↗

arXiv2021

Hyperspectral Image Super-Resolution in Arbitrary Input-Output Band Settings

Hyperspectral image (HSI) with narrow spectral bands can capture rich spectral information, but it sacrifices its spatial resolution in the process. Many machine-learning-based HSI super-resolution (SR) algorithms have been proposed recently. However, one of the fundamental limitations of these approaches is that they are highly dependent on image and camera settings and can only learn to map an input HSI with one specific setting to an output HSI with another. However, different cameras capture images with different spectral response functions and bands numbers due to the diversity of HSI cameras. Consequently, the existing machine-learning-based approaches fail to learn to super-resolve HSIs for a wide variety of input-output band settings. We propose a single Meta-Learning-Based Super-Resolution (MLSR) model, which can take in HSI images at an arbitrary number of input bands' peak wavelengths and generate SR HSIs with an arbitrary number of output bands' peak wavelengths. We leverage NTIRE2020 and ICVL datasets to train and validate the performance of the MLSR model. The results show that the single proposed model can successfully generate super-resolved HSI bands at arbitrary input-output band settings. The results are better or at least comparable to baselines that are separately trained on a specific input-output band setting.

cs.CV↗

arXiv2018

Solving Poisson's Equation using Deep Learning in Particle Simulation of PN Junction

Simulating the dynamic characteristics of a PN junction at the microscopic level requires solving the Poisson's equation at every time step. Solving at every time step is a necessary but time-consuming process when using the traditional finite difference (FDM) approach. Deep learning is a powerful technique to fit complex functions. In this work, deep learning is utilized to accelerate solving Poisson's equation in a PN junction. The role of the boundary condition is emphasized in the loss function to ensure a better fitting. The resulting I-V curve for the PN junction, using the deep learning solver presented in this work, shows a perfect match to the I-V curve obtained using the finite difference method, with the advantage of being 10 times faster at every time step.

physics.comp-ph↗

arXiv2013

Quasinormal ringing of Kerr black holes. II. Excitation by particles falling radially with arbitrary energy

The analytical understanding of quasinormal mode ringing requires an accurate knowledge of the Green's function describing the response of the black hole to external perturbations. We carry out a comprehensive study of quasinormal mode excitation for Kerr black holes. Relying on the formalism developed by Mano, Suzuki and Takasugi, we improve and extend previous calculations of the quasinormal mode residues in the complex frequency plane ("excitation factors" B_q). Using these results we compute the "excitation coefficients" C_q (essentially the mode amplitudes) in the special case where the source of the perturbations is a particle falling into the black hole along the symmetry axis. We compare this calculation with numerical integrations of the perturbation equations, and we show quantitatively how the addition of higher overtones improves the agreement with the numerical waveforms. Our results should find applications in models of the ringdown stage and in the construction of semianalytical template banks for gravitational-wave detectors, especially for binaries with large mass ratios and/or fast-spinning black holes.

gr-qc↗

arXiv2012

Quasinormal-mode spectrum of Kerr black holes and its geometric interpretation

There is a well-known, intuitive geometric correspondence between high-frequency QNMs of Schwarzschild black holes and null geodesics that reside on the light-ring : the real part of the mode's frequency relates to the geodesic's orbital frequency, and the imaginary part of the frequency corresponds to the Lyapunov exponent of the orbit. For slowly rotating black holes, the QNM real frequency is a linear combination of a the orbit's precessional and orbital frequencies, but the correspondence is otherwise unchanged. In this paper, we find a relationship between the QNM frequencies of Kerr black holes of arbitrary (astrophysical) spins and general spherical photon orbits, which is analogous to the relationship for slowly rotating holes. To derive this result, we first use the WKB approximation to compute accurate algebraic expressions for large-l QNM frequencies. Comparing our WKB calculation to the leading-order, geometric-optics approximation to scalar-wave propagation in the Kerr spacetime, we then draw a correspondence between the real parts of the parameters of a QNM and the conserved quantities of spherical photon orbits. At next-to-leading order in this comparison, we relate the imaginary parts of the QNM parameters to coefficients that modify the amplitude of the scalar wave. With this correspondence, we find a geometric interpretation to two features of the QNM spectrum of Kerr black holes: First, for Kerr holes rotating near the maximal rate, a large number of modes have nearly zero damping; we connect this characteristic to the fact that a large number of spherical photon orbits approach the horizon in this limit. Second, for black holes of any spins, the frequencies of specific sets of modes are degenerate; we find that this feature arises when the spherical photon orbits corresponding to these modes form closed (as opposed to ergodically winding) curves.

gr-qc↗

arXiv2012

Reconstructing DNA copy number by joint segmentation of multiple sequences

The variation in DNA copy number carries information on the modalities of genome evolution and misregulation of DNA replication in cancer cells; its study can be helpful to localize tumor suppressor genes, distinguish different populations of cancerous cell, as well identify genomic variations responsible for disease phenotypes. A number of different high throughput technologies can be used to identify copy number variable sites, and the literature documents multiple effective algorithms. We focus here on the specific problem of detecting regions where variation in copy number is relatively common in the sample at hand: this encompasses the cases of copy number polymorphisms, related samples, technical replicates, and cancerous sub-populations from the same individual. We present an algorithm based on regularization approaches with significant computational advantages and competitive accuracy. We illustrate its applicability with simulated and real data sets.

stat.ME↗

arXiv2011

Accuracy of the post-Newtonian approximation. II. Optimal asymptotic expansion of the energy flux for quasicircular, extreme mass-ratio inspirals into a Kerr black hole

We study the effect of black hole spin on the accuracy of the post-Newtonian approximation. We focus on the gravitational energy flux for the quasicircular, equatorial, extreme mass-ratio inspiral of a compact object into a Kerr black hole of mass M and spin J. For a given dimensionless spin a=J/M^2 (in geometrical units), the energy flux depends only on the orbital velocity v or (equivalently) on the Boyer-Lindquist orbital radius r. We investigate the formal region of validity of the Taylor post-Newtonian expansion of the energy flux (which is known up to order v^8 beyond the quadrupole formula), generalizing previous work by two of us. The "error function" used to determine the region of validity of the post-Newtonian expansion can have two qualitatively different kinds of behavior, and we deal with these two cases separately. We find that, at any fixed post-Newtonian order, the edge of the region of validity (as measured by v/v_{ISCO}, where v_{ISCO} is the orbital velocity at the innermost stable circular orbit) is only weakly dependent on a. Unlike in the nonspinning case, the lack of sufficiently high order terms does not allow us to determine if there is a convergent to divergent transition at order v^6. Independently of a, the inclusion of angular multipoles up to and including l=5 in the numerical flux is necessary to achieve the level of accuracy of the best-known (N=8) PN expansion of the energy flux.

gr-qc↗

arXiv2011

Reconstructing DNA copy number by penalized estimation and imputation

Recent advances in genomics have underscored the surprising ubiquity of DNA copy number variation (CNV). Fortunately, modern genotyping platforms also detect CNVs with fairly high reliability. Hidden Markov models and algorithms have played a dominant role in the interpretation of CNV data. Here we explore CNV reconstruction via estimation with a fused-lasso penalty as suggested by Tibshirani and Wang [Biostatistics 9 (2008) 18--29]. We mount a fresh attack on this difficult optimization problem by the following: (a) changing the penalty terms slightly by substituting a smooth approximation to the absolute value function, (b) designing and implementing a new MM (majorization--minimization) algorithm, and (c) applying a fast version of Newton's method to jointly update all model parameters. Together these changes enable us to minimize the fused-lasso criterion in a highly effective way. We also reframe the reconstruction problem in terms of imputation via discrete optimization. This approach is easier and more accurate than parameter estimation because it relies on the fact that only a handful of possible copy number states exist at each SNP. The dynamic programming framework has the added bonus of exploiting information that the current fused-lasso approach ignores. The accuracy of our imputations is comparable to that of hidden Markov models at a substantially lower computational cost.

stat.ME↗