BZPEERReading, trust and collaboration for research
Menu

Discover

GraphFollow connections around a paper, topic or saved view.

Workspaces

HomePick up where your research flow left off.InboxHandle requests, replies and notifications from one place.CollectionsCurate reading lists, evidence sets and shareable dossiers.ResearchSee your private provenance graph, trail and imports.CollaborationMove from discovery into focused discussion rooms.PublishDraft, preview and publish full-text research articles.

Network

ExpertsFind specialists worth following or asking for help.CommunitiesJoin research circles organized around shared work.PartnersSee external organizations active around the same topics.

Opportunities

ListingsBrowse roles, calls and collaborations with clearer fit signals.

Account

Log inReturn to your reading and collaboration workspace.Create accountStart saving work, following people and joining teams.
Log inCreate account

Researcher profile

Qian Zheng

Qian Zheng contributes to research discovery and scholarly infrastructure.

ResearcherAffiliation not importedOpen to collaborate
astro-ph.COComputer Visionastro-ph.GAastro-ph.IMGraphicsMachine Learningmath.OC

Trust snapshot

Quick read

Trust 21 - Emerging
14works
0followers
7topics
4close collaborators

Actions

Decide how to stay connected

Follow researcher0

Research graph

See the researcher in context

Open full explorer

Inspect adjacent work, topics, institutions and collaborators without jumping out to a separate graph page.

Building this graph slice

BZPEER is loading the nearby papers, people, topics and institutions for this page.

Published work

14 published item(s)

preprint2026arXiv

Neural-Driven Image Editing

Traditional image editing typically relies on manual prompting, making it labor-intensive and inaccessible to individuals with limited motor control or language abilities. Leveraging recent advances in brain-computer interfaces (BCIs) and generative models, we propose LoongX, a hands-free image editing approach driven by multimodal neurophysiological signals. LoongX utilizes state-of-the-art diffusion models trained on a comprehensive dataset of 23,928 image editing pairs, each paired with synchronized electroencephalography (EEG), functional near-infrared spectroscopy (fNIRS), photoplethysmography (PPG), and head motion signals that capture user intent. To effectively address the heterogeneity of these signals, LoongX integrates two key modules. The cross-scale state space (CS3) module encodes informative modality-specific features. The dynamic gated fusion (DGF) module further aggregates these features into a unified latent space, which is then aligned with edit semantics via fine-tuning on a diffusion transformer (DiT). Additionally, we pre-train the encoders using contrastive learning to align cognitive states with semantic intentions from embedded natural language. Extensive experiments demonstrate that LoongX achieves performance comparable to text-driven methods (CLIP-I: 0.6605 vs. 0.6558; DINO: 0.4812 vs. 0.4636) and outperforms them when neural signals are combined with speech (CLIP-T: 0.2588 vs. 0.2549). These results highlight the promise of neural-driven generative models in enabling accessible, intuitive image editing and open new directions for cognitive-driven creative technologies. The code and dataset are released on the project website: https://loongx1.github.io.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Automatic Facial Skin Feature Detection for Everyone

Automatic assessment and understanding of facial skin condition have several applications, including the early detection of underlying health problems, lifestyle and dietary treatment, skin-care product recommendation, etc. Selfies in the wild serve as an excellent data resource to democratize skin quality assessment, but suffer from several data collection challenges.The key to guaranteeing an accurate assessment is accurate detection of different skin features. We present an automatic facial skin feature detection method that works across a variety of skin tones and age groups for selfies in the wild. To be specific, we annotate the locations of acne, pigmentation, and wrinkle for selfie images with different skin tone colors, severity levels, and lighting conditions. The annotation is conducted in a two-phase scheme with the help of a dermatologist to train volunteers for annotation. We employ Unet++ as the network architecture for feature detection. This work shows that the two-phase annotation scheme can robustly detect the accurate locations of acne, pigmentation, and wrinkle for selfie images with different ethnicities, skin tone colors, severity levels, age groups, and lighting conditions.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Impact of the turnover in the high-z galaxy luminosity function on the 21-cm signal during Cosmic Dawn and Epoch of Reionization

The shape of the faint-end of the high-z galaxy luminosity function (LF) informs early star formation and reionization physics during the Cosmic Dawn and Epoch of Reionization. Until recently, based on the strong gravitational lensing cluster deep surveys, the Hubble Frontier Fields (HFF) has found a potential turnover in the ultraviolet (UV) LF at z$\sim$6. In this paper, we analyze the contribution of extremely faint galaxies with the magnitude larger than the turnover magnitude in LF to cosmic reionization. We apply the measurement from HFF to our suppressed star formation efficiency model, including three free parameters: halo mass threshold $M_t$, curvature parameter $β$ and a UV conversion factor $l_{\rm UV}$. According to our fit of 68\% confidence level, the high-redshift star formation in haloes smaller than $ M_t=1.82^{+2.86}_{-1.08}\times10^{10} \rm M_{\odot}$ is found to be dampened. The turnover magnitude $\rm \gtrsim -13.99-2.45$, correspondingly the halo mass $\lesssim(4.57+20.03)\times10^{9} \rm M_{\odot}$. We find that the absorption trough in the global 21-cm signal is sensitive to our SFE model parameters. Together with ($β$, $l_{\rm UV}$) = ($2.17^{+2.42}_{-1.72}$, $9.33^{+0.43}_{-0.42} \rm ~erg~yr ~s^{-1}M_{\odot}^{-1})$, the trough locates at $\sim$ $134^{+10}_{-17}$ $\rm MHz$ with an amplitude of $\sim$ $-237^{-6}_{+7}$ $\rm mK$, compared to (106\rm MHz, -212\rm mK) in the absence of turnover. Besides, we find that the star formation of faint galaxies has also an impact on the 21-cm power spectra. The best fitting peak power decreases by $\sim4\%$ and shifts towards smaller scales from $0.88 h \rm Mpc^{-1}$ to $0.91 h \rm Mpc^{-1}$. According to our calculation, such impact is distinguishable with the forthcoming Square Kilometre Array.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

NeIF: Representing General Reflectance as Neural Intrinsics Fields for Uncalibrated Photometric Stereo

Uncalibrated photometric stereo (UPS) is challenging due to the inherent ambiguity brought by unknown light. Existing solutions alleviate the ambiguity by either explicitly associating reflectance to light conditions or resolving light conditions in a supervised manner. This paper establishes an implicit relation between light clues and light estimation and solves UPS in an unsupervised manner. The key idea is to represent the reflectance as four neural intrinsics fields, i.e., position, light, specular, and shadow, based on which the neural light field is implicitly associated with light clues of specular reflectance and cast shadow. The unsupervised, joint optimization of neural intrinsics fields can be free from training data bias as well as accumulating error, and fully exploits all observed pixel values for UPS. Our method achieves a superior performance advantage over state-of-the-art UPS methods on public and self-collected datasets, under regular and challenging setups. The code will be released soon.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Policy Optimization with Stochastic Mirror Descent

Improving sample efficiency has been a longstanding goal in reinforcement learning. This paper proposes $\mathtt{VRMPO}$ algorithm: a sample efficient policy gradient method with stochastic mirror descent. In $\mathtt{VRMPO}$, a novel variance-reduced policy gradient estimator is presented to improve sample efficiency. We prove that the proposed $\mathtt{VRMPO}$ needs only $\mathcal{O}(ε^{-3})$ sample trajectories to achieve an $ε$-approximate first-order stationary point, which matches the best sample complexity for policy optimization. The extensive experimental results demonstrate that $\mathtt{VRMPO}$ outperforms the state-of-the-art policy gradient methods in various settings.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Object Properties Inferring from and Transfer for Human Interaction Motions

Humans regularly interact with their surrounding objects. Such interactions often result in strongly correlated motion between humans and the interacting objects. We thus ask: "Is it possible to infer object properties from skeletal motion alone, even without seeing the interacting object itself?" In this paper, we present a fine-grained action recognition method that learns to infer such latent object properties from human interaction motion alone. This inference allows us to disentangle the motion from the object property and transfer object properties to a given motion. We collected a large number of videos and 3D skeletal motions of the performing actors using an inertial motion capture device. We analyze similar actions and learn subtle differences among them to reveal latent properties of the interacting objects. In particular, we learn to identify the interacting object, by estimating its weight, or its fragility or delicacy. Our results clearly demonstrate that the interaction motions and interacting objects are highly correlated and indeed relative object latent properties can be inferred from the 3D skeleton sequences alone, leading to new synthesis possibilities for human interaction motions. Dataset will be available at http://vcc.szu.edu.cn/research/2020/IT.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Searching for Dark Matter Signals from Local Dwarf Spheroidal Galaxies at Low Radio Frequencies in the GLEAM Survey

The search for emission from weakly interacting massive particle (WIMP) dark matter annihilation and decay has become a multi-pronged area of research not only targeting a diverse selection of astrophysical objects, but also taking advantage of the entire electromagnetic spectrum. The decay of WIMP particles into standard model particles has been suggested as a possible channel for synchrotron emission to be detected at low radio frequencies. Here, we present the stacking analysis of a sample of 33 dwarf spheroidal (dSph) galaxies with low-frequency (72 - 231 MHz) radio images from the GaLactic and Extragalactic All-sky Murchison Widefield Array (GLEAM) survey. We produce radial surface brightness profiles of images centred upon each dSph galaxy with background radio sources masked. We remove ten fields from the stacking due to contamination from either poorly subtracted, bright radio sources or strong background gradients across the field. The remaining 23 dSph galaxies are stacked in an attempt to obtain a statistical detection of any WIMP-induced synchrotron emission in these systems. We find that the stacked radial brightness profile does not exhibit a statistically significant detection above the 95% confidence level of $\sim$1.5 mJy beam$^{-1}$. This novel technique shows the potential of using low-frequency radio images to constrain fundamental properties of particle dark matter.

0 citations0 reviewsNo code linkedOpen access
preprint2016arXiv

Radio Sources in the NCP Region Observed with the 21 Centimeter Array

We present a catalog of 624 radio sources detected around the North Celestial Pole (NCP) with the 21 Centimeter Array (21CMA), a radio interferometer dedicated to the statistical measurement of the epoch of reionization (EoR). The data are taken from a 12 h observation made on 2013 April 13, with a frequency coverage from 75 to 175 MHz and an angular resolution of ~ 4 arcmin. The catalog includes flux densities at eight sub-bands across the 21CMA bandwidth and provides the in-band spectral indices for the detected sources. To reduce the complexity of interferometric imaging from the so-called "w" term and ionospheric effects, the present analysis are restricted to the east-west baselines within 1500 m only. The 624 radio sources are found within 5 degrees around the NCP down to ~ 0.1 Jy. Our source counts are compared, and also exhibit a good agreement, with deep low-frequency observations made recently with the GMRT and MWA. In particular, for fainter radio sources below ~ 1 Jy, we find a flattening trend of source counts towards lower frequencies. While the thermal noise (~0.4 mJy) is well controlled to below the confusion limit, the dynamical range (~10^4) and sensitivity of current 21CMA imaging is largely limited by calibration and deconvolution errors, especially the grating lobes of very bright sources, such as 3C061.1, in the NCP field which result from the regular spacings of the 21CMA. We note that particular attention should be paid to the extended sources, and their modeling and removals may constitute a large technical challenge for current EoR experiments. Our analysis may serve as a useful guide to design of next generation low-frequency interferometers like the Square Kilometre Array.

0 citations0 reviewsNo code linkedOpen access
preprint2016arXiv

The radio environment of the 21 Centimeter Array: RFI detection and mitigation

Detection and mitigation of radio frequency interference (RFI) is the first and also the key step for data processing in radio observations, especially for ongoing low frequency radio experiments towards the detection of the cosmic dawn and epoch of reionization (EoR). In this paper we demonstrate the technique and efficiency of RFI identification and mitigation for the 21 Centimeter Array (21CMA), a radio interferometer dedicated to the statistical measurement of EoR. For terrestrial, man-made RFI, we concentrate mainly on a statistical approach by identifying and then excising non-Gaussian signatures, in the sense that the extremely weak cosmic signal is actually buried under thermal and therefore Gaussian noise. We also introduce the so-called visibility correlation coefficient instead of conventional visibility, which allows a further suppression of rapidly time-varying RFI. Finally, we briefly discuss removals of the sky RFI, the leakage of sidelobes from off-field strong radio sources with time-invariant power and a featureless spectrum. It turns out that state of the art technique should allow us to detect and mitigate RFI to a satisfactory level in present low frequency interferometer observations such as those acquired with the 21CMA, and the accuracy and efficiency can be greatly improved with the employment of low-cost, high-speed computing facilities for data acquisition and processing.

0 citations0 reviewsNo code linkedOpen access
preprint2013arXiv

Simplification Methods for Sum-of-Squares Programs

A sum-of-squares is a polynomial that can be expressed as a sum of squares of other polynomials. Determining if a sum-of-squares decomposition exists for a given polynomial is equivalent to a linear matrix inequality feasibility problem. The computation required to solve the feasibility problem depends on the number of monomials used in the decomposition. The Newton polytope is a method to prune unnecessary monomials from the decomposition. This method requires the construction of a convex hull and this can be time consuming for polynomials with many terms. This paper presents a new algorithm for removing monomials based on a simple property of positive semidefinite matrices. It returns a set of monomials that is never larger than the set returned by the Newton polytope method and, for some polynomials, is a strictly smaller set. Moreover, the algorithm takes significantly less computation than the convex hull construction. This algorithm is then extended to a more general simplification method for sum-of-squares programming.

0 citations0 reviewsNo code linkedOpen access
preprint2012arXiv

A Method to Extract the Angular Power Spectrum of the Epoch of Reionization from Low-Frequency Radio Interferometers

The redshifted 21cm signal of neutral hydrogen from the epoch of reionization (EoR) is extremely weak and its first detection is therefore expected to be statistical with first-generation low-frequency radio interferometers. In this letter we propose a method to extract the angular power spectrum of EoR from the visibility correlation coefficients p_{ij}(u,v), instead of the visibilities V_{ij}(u,v) measured directly by radio interferometers in conventional algorithm. The visibility correlation coefficients are defined as p_{ij}(u,v)=V_{ij}(u,v)/\sqrt{|V_{ii}||V_{jj}|} by introducing the auto-correlation terms V_{ii} and V_{jj} such that the angular power spectrum C_{\ell} can be obtained through C_{\ell}=4pi^2T_0^2<|p_{ij}(u,v)|^2>, independently of the primary beams of antennas. This also removes partially the influence of receiver gains in the measurement of C_{\ell} because the amplitudes of the gains cancel each other out in the statistical average operation of <|p_{ij}(u,v)|^2>.We use the average system temperature T_0 as a calibrator of C_{\ell}, which is dominated by the Milky Way and extragalactic sources in our interested frequency range below 200 MHz. Finally we demonstrate the feasibility of the novel method using the simulated sky maps as targets and the 21 CentiMeter Array (21CMA) as interferometer.

0 citations0 reviewsNo code linkedOpen access
preprint2012arXiv

Influence of synchrotron self-absorption on the 21cm experiments

Presence of spectral curvature due to synchrotron self-absorption (SSA) of extragalactic radio sources may break down the spectral smoothness feature-the premise that bright radio foreground can be successfully removed in the 21cm experiments of searching for the epoch of reionization (EOR).We present a quantitative estimate of the effect on the measurement of the angular power spectrum of the low-frequency sky,incorporating a phenomenological model,characterized by the fraction of radio sources with turnover frequencies in 100-1000MHz range and a broken power law for the spectral transition around turnover frequencies nu_m,into the simulated radio sources over a small sky area of 10x10 deg^2.We compare statistically the changes in their residual maps with/without inclusion of the SSA after the bright sources of S_150MHz>=100mJy are excised and the best-fitted polynomials in frequency domain on each pixel are further subtracted.It has been shown that the effect of SSA on the detection of EOR depends sensitively on the spectral profiles of radio sources around the turnover frequencies: A hard transition model described by the broken power law with the turnover of spectral index at nu_m would leave pronounced imprints on the residual background and cause serious confusion with the EOR signal.However,the spectral signatures on the angular power spectrum of extragalactic foreground generated by a soft transition model,in which the rise and fall power laws of spectral distribution around nu_m are connected through a smooth transition spanning >=200 MHz in characteristic width,can be fitted and consequently subtracted by employment of polynomials to an acceptable degree(delta T<1mK).As this latter scenario seems to be favored by both theoretical expectation and radio spectral observations,we conclude that the influence of SSA on the 21cm experiments is probably very minor.

0 citations0 reviewsNo code linkedOpen access
preprint2010arXiv

How to Identify and Separate Bright Galaxy Clusters from the Low-frequency Radio Sky?

In this work we simulate the $50-200$ MHz radio sky that is constrained in the field of view ($5^{\circ}$ radius) of the 21 Centimeter Array (21CMA), by carrying out Monte-Carlo simulations to model redshifted cosmological reionization signals and strong contaminating foregrounds, including emissions from our Galaxy, galaxy clusters, and extragalactic point sources. As an improvement of previous works, we consider in detail not only random variations of morphological and spectroscopic parameters within the ranges allowed by multi-band observations, but also evolution of radio halos in galaxy clusters, assuming that relativistic electrons are re-accelerated in the ICM in merger events and lose energy via both synchrotron emission and inverse Compton scattering with CMB photons. By introducing a new approach designed on the basis of independent component analysis (ICA) and wavelet detection algorithm, we prove that, with a cumulative observation of one month with the 21CMA array, about $80\%$ of galaxy clusters with central brightness temperatures of $> 10~{\rm K}$ at 65 MHz can be safely identified and separated from the overwhelmingly bright foreground. We find that the morphological and spectroscopic distortions are extremely small as compared to the input simulated clusters, and the reduced $χ^2$ of brightness temperature profiles and spectra are controlled to be $\lesssim 0.5$ and $\lesssim 1.3$, respectively. These results robustly indicate that in the near future a sample of dozens of bright galaxy clusters will be disentangled from the foreground in 21CMA observations, the study of which will greatly improve our knowledge about cluster merger rates, electron acceleration mechanisms in cluster radio halos, and magnetic field in the ICM.

0 citations0 reviewsNo code linkedOpen access
preprint2009arXiv

Contribution of Cross-Correlations to the 21cm Angular Power Spectrum in the Epoch of Reionization

Measurement of the 21cm hyperfine transition of neutral hydrogen provides a unique probe of the epoch of reionization and the Dark Ages. Three major mechanisms are believed to dominate the radiation process: emission from neutral hydrogen surrounding the ionized bubbles of first galaxies and/or quasars, emission from neutral hydrogen inside minihalos, and absorption of diffuse neutral hydrogen against the cosmic microwave background. In the present work, by simply combining the existing analytic models for the three mechanisms, we investigate the contribution of cross-correlation between these three components to the total 21cm angular power spectrum, in the sense that neutral hydrogen associated with different radiation processes traces the large-scale structures of underlying density perturbations. While the overall 21cm power spectrum remains almost unchanged with the inclusion of the cross-correlations, the cross-correlation may play a key role in the determination of the 21cm power spectrum during the transition of 21cm radiation from emission-dominated phase to absorption-dominated phase at redshift z~20. A significant suppression in the 21cm angular power spectrum during this transition is anticipated as the result of negative contribution of the cross-correlation between the absorption of diffuse neutral hydrogen and the emission components. Therefore, an accurate prediction of the cosmic 21cm power spectrum should take the cross-correlation into account especially at the transition phase.

0 citations0 reviewsNo code linkedOpen access