arXiv subjects

Zhang Chen

Publications and source records attributed to Zhang Chen.

At least 55 records · Page 3Linked to original sources

arXiv2023

Relit-NeuLF: Efficient Relighting and Novel View Synthesis via Neural 4D Light Field

In this paper, we address the problem of simultaneous relighting and novel view synthesis of a complex scene from multi-view images with a limited number of light sources. We propose an analysis-synthesis approach called Relit-NeuLF. Following the recent neural 4D light field network (NeuLF), Relit-NeuLF first leverages a two-plane light field representation to parameterize each ray in a 4D coordinate system, enabling efficient learning and inference. Then, we recover the spatially-varying bidirectional reflectance distribution function (SVBRDF) of a 3D scene in a self-supervised manner. A DecomposeNet learns to map each ray to its SVBRDF components: albedo, normal, and roughness. Based on the decomposed BRDF components and conditioning light directions, a RenderNet learns to synthesize the color of the ray. To self-supervise the SVBRDF decomposition, we encourage the predicted ray color to be close to the physically-based rendering result using the microfacet model. Comprehensive experiments demonstrate that the proposed method is efficient and effective on both synthetic data and real-world human face data, and outperforms the state-of-the-art results. We publicly released our code on GitHub. You can find it here: https://github.com/oppo-us-research/RelitNeuLF

cs.CV↗

arXiv2023

NeuRBF: A Neural Fields Representation with Adaptive Radial Basis Functions

We present a novel type of neural fields that uses general radial bases for signal representation. State-of-the-art neural fields typically rely on grid-based representations for storing local neural features and N-dimensional linear kernels for interpolating features at continuous query points. The spatial positions of their neural features are fixed on grid nodes and cannot well adapt to target signals. Our method instead builds upon general radial bases with flexible kernel position and shape, which have higher spatial adaptivity and can more closely fit target signals. To further improve the channel-wise capacity of radial basis functions, we propose to compose them with multi-frequency sinusoid functions. This technique extends a radial basis to multiple Fourier radial bases of different frequency bands without requiring extra parameters, facilitating the representation of details. Moreover, by marrying adaptive radial bases with grid-based ones, our hybrid combination inherits both adaptivity and interpolation smoothness. We carefully designed weighting schemes to let radial bases adapt to different types of signals effectively. Our experiments on 2D image and 3D signed distance field representation demonstrate the higher accuracy and compactness of our method than prior arts. When applied to neural radiance field reconstruction, our method achieves state-of-the-art rendering quality, with small model size and comparable training speed.

cs.CV↗

arXiv2023

High Fidelity 3D Hand Shape Reconstruction via Scalable Graph Frequency Decomposition

Despite the impressive performance obtained by recent single-image hand modeling techniques, they lack the capability to capture sufficient details of the 3D hand mesh. This deficiency greatly limits their applications when high-fidelity hand modeling is required, e.g., personalized hand modeling. To address this problem, we design a frequency split network to generate 3D hand mesh using different frequency bands in a coarse-to-fine manner. To capture high-frequency personalized details, we transform the 3D mesh into the frequency domain, and propose a novel frequency decomposition loss to supervise each frequency component. By leveraging such a coarse-to-fine scheme, hand details that correspond to the higher frequency domain can be preserved. In addition, the proposed network is scalable, and can stop the inference at any resolution level to accommodate different hardware with varying computational powers. To quantitatively evaluate the performance of our method in terms of recovering personalized shape details, we introduce a new evaluation metric named Mean Signal-to-Noise Ratio (MSNR) to measure the signal-to-noise ratio of each mesh frequency component. Extensive experiments demonstrate that our approach generates fine-grained details for high-fidelity 3D hand reconstruction, and our evaluation metric is more effective for measuring mesh details compared with traditional metrics.

cs.CV↗

arXiv2023

Harnessing Low-Frequency Neural Fields for Few-Shot View Synthesis

Neural Radiance Fields (NeRF) have led to breakthroughs in the novel view synthesis problem. Positional Encoding (P.E.) is a critical factor that brings the impressive performance of NeRF, where low-dimensional coordinates are mapped to high-dimensional space to better recover scene details. However, blindly increasing the frequency of P.E. leads to overfitting when the reconstruction problem is highly underconstrained, \eg, few-shot images for training. We harness low-frequency neural fields to regularize high-frequency neural fields from overfitting to better address the problem of few-shot view synthesis. We propose reconstructing with a low-frequency only field and then finishing details with a high-frequency equipped field. Unlike most existing solutions that regularize the output space (\ie, rendered images), our regularization is conducted in the input space (\ie, signal frequency). We further propose a simple-yet-effective strategy for tuning the frequency to avoid overfitting few-shot inputs: enforcing consistency among the frequency domain of rendered 2D images. Thanks to the input space regularizing scheme, our method readily applies to inputs beyond spatial locations, such as the time dimension in dynamic scenes. Comparisons with state-of-the-art on both synthetic and natural datasets validate the effectiveness of our proposed solution for few-shot view synthesis. Code is available at \href{https://github.com/lsongx/halo}{https://github.com/lsongx/halo}.

cs.CV↗

arXiv2023

NeRFPlayer: A Streamable Dynamic Scene Representation with Decomposed Neural Radiance Fields

Visually exploring in a real-world 4D spatiotemporal space freely in VR has been a long-term quest. The task is especially appealing when only a few or even single RGB cameras are used for capturing the dynamic scene. To this end, we present an efficient framework capable of fast reconstruction, compact modeling, and streamable rendering. First, we propose to decompose the 4D spatiotemporal space according to temporal characteristics. Points in the 4D space are associated with probabilities of belonging to three categories: static, deforming, and new areas. Each area is represented and regularized by a separate neural field. Second, we propose a hybrid representations based feature streaming scheme for efficiently modeling the neural fields. Our approach, coined NeRFPlayer, is evaluated on dynamic scenes captured by single hand-held cameras and multi-camera arrays, achieving comparable or superior rendering performance in terms of quality and speed comparable to recent state-of-the-art methods, achieving reconstruction in 10 seconds per frame and interactive rendering.

cs.CV↗

arXiv2022

A data-enabled physics-informed neural network with comprehensive numerical study on solving neutron diffusion eigenvalue problems

We present a data-enabled physics-informed neural network (DEPINN) with comprehensive numerical study for solving industrial scale neutron diffusion eigenvalue problems (NDEPs). In order to achieve an engineering acceptable accuracy for complex engineering problems, a very small amount of prior data from physical experiments are suggested to be used, to improve the accuracy and efficiency of training. We design an adaptive optimization procedure with Adam and LBFGS to accelerate the convergence in the training stage. We discuss the effect of different physical parameters, sampling techniques, loss function allocation and the generalization performance of the proposed DEPINN model for solving complex problem. The feasibility of proposed DEPINN model is tested on three typical benchmark problems, from simple geometry to complex geometry, and from mono-energetic equation to two-group equations. Numerous numerical results show that DEPINN can efficiently solve NDEPs with an appropriate optimization procedure. The proposed DEPINN can be generalized for other input parameter settings once its structure been trained. This work confirms the possibility of DEPINN for practical engineering applications in nuclear reactor physics.

physics.comp-ph↗

arXiv2022

Prospects for Detecting the Diffuse Supernova Neutrino Background with JUNO

We present the detection potential for the diffuse supernova neutrino background (DSNB) at the Jiangmen Underground Neutrino Observatory (JUNO), using the inverse-beta-decay (IBD) detection channel on free protons. We employ the latest information on the DSNB flux predictions, and investigate in detail the background and its reduction for the DSNB search at JUNO. The atmospheric neutrino induced neutral current (NC) background turns out to be the most critical background, whose uncertainty is carefully evaluated from both the spread of model predictions and an envisaged \textit{in situ} measurement. We also make a careful study on the background suppression with the pulse shape discrimination (PSD) and triple coincidence (TC) cuts. With latest DSNB signal predictions, more realistic background evaluation and PSD efficiency optimization, and additional TC cut, JUNO can reach the significance of 3$σ$ for 3 years of data taking, and achieve better than 5$σ$ after 10 years for a reference DSNB model. In the pessimistic scenario of non-observation, JUNO would strongly improve the limits and exclude a significant region of the model parameter space.

hep-ex↗

arXiv2022

Damping signatures at JUNO, a medium-baseline reactor neutrino oscillation experiment

We study damping signatures at the Jiangmen Underground Neutrino Observatory (JUNO), a medium-baseline reactor neutrino oscillation experiment. These damping signatures are motivated by various new physics models, including quantum decoherence, $ν_3$ decay, neutrino absorption, and wave packet decoherence. The phenomenological effects of these models can be characterized by exponential damping factors at the probability level. We assess how well JUNO can constrain these damping parameters and how to disentangle these different damping signatures at JUNO. Compared to current experimental limits, JUNO can significantly improve the limits on $τ_3/m_3$ in the $ν_3$ decay model, the width of the neutrino wave packet $σ_x$, and the intrinsic relative dispersion of neutrino momentum $σ_{\rm rel}$.

hep-ex↗

arXiv2022

Light Field-Based Underwater 3D Reconstruction Via Angular Resampling

Recovering 3D geometry of underwater scenes is challenging because of non-linear refraction of light at the water-air interface caused by the camera housing. We present a light field-based approach that leverages properties of angular samples for high-quality underwater 3D reconstruction from a single viewpoint. Specifically, we resample the light field image to angular patches. As underwater scenes exhibit weak view-dependent specularity, an angular patch tends to have uniform intensity when sampled at the correct depth. We thus impose this angular uniformity as a constraint for depth estimation. For efficient angular resampling, we design a fast approximation algorithm based on multivariate polynomial regression to approximate nonlinear refraction paths. We further develop a light field calibration algorithm that estimates the water-air interface geometry along with the camera parameters. Comprehensive experiments on synthetic and real data show our method produces state-of-the-art reconstruction on static and dynamic underwater scenes.

cs.CV↗

arXiv2021

Radioactivity control strategy for the JUNO detector

JUNO is a massive liquid scintillator detector with a primary scientific goal of determining the neutrino mass ordering by studying the oscillated anti-neutrino flux coming from two nuclear power plants at 53 km distance. The expected signal anti-neutrino interaction rate is only 60 counts per day, therefore a careful control of the background sources due to radioactivity is critical. In particular, natural radioactivity present in all materials and in the environment represents a serious issue that could impair the sensitivity of the experiment if appropriate countermeasures were not foreseen. In this paper we discuss the background reduction strategies undertaken by the JUNO collaboration to reduce at minimum the impact of natural radioactivity. We describe our efforts for an optimized experimental design, a careful material screening and accurate detector production handling, and a constant control of the expected results through a meticulous Monte Carlo simulation program. We show that all these actions should allow us to keep the background count rate safely below the target value of 10 Hz in the default fiducial volume, above an energy threshold of 0.7 MeV.

physics.ins-det↗

arXiv2021

JUNO sensitivity to low energy atmospheric neutrino spectra

Atmospheric neutrinos are one of the most relevant natural neutrino sources that can be exploited to infer properties about cosmic rays and neutrino oscillations. The Jiangmen Underground Neutrino Observatory (JUNO) experiment, a 20 kton liquid scintillator detector with excellent energy resolution is currently under construction in China. JUNO will be able to detect several atmospheric neutrinos per day given the large volume. A study on the JUNO detection and reconstruction capabilities of atmospheric $ν_e$ and $ν_μ$ fluxes is presented in this paper. In this study, a sample of atmospheric neutrino Monte Carlo events has been generated, starting from theoretical models, and then processed by the detector simulation. The excellent timing resolution of the 3'' PMT light detection system of JUNO detector and the much higher light yield for scintillation over Cherenkov allow to measure the time structure of the scintillation light with very high precision. Since $ν_e$ and $ν_μ$ interactions produce a slightly different light pattern, the different time evolution of light allows to discriminate the flavor of primary neutrinos. A probabilistic unfolding method has been used, in order to infer the primary neutrino energy spectrum from the detector experimental observables. The simulated spectrum has been reconstructed between 100 MeV and 10 GeV, showing a great potential of the detector in the atmospheric low energy region.

hep-ex↗

arXiv2021

Multiple-Pilot Collaboration for Advanced Remote Intervention using Reinforcement Learning

The traditional master-slave teleoperation relies on human expertise without correction mechanisms, resulting in excessive physical and mental workloads. To address these issues, a co-pilot-in-the-loop control framework is investigated for cooperative teleoperation. A deep deterministic policy gradient(DDPG) based agent is realised to effectively restore the master operators' intents without prior knowledge on time delay. The proposed framework allows for introducing an operator (i.e., co-pilot) to generate commands at the slave side, whose weights are optimally assigned online through DDPG-based arbitration, thereby enhancing the command robustness in the case of possible human operational errors. With the help of interval type-2(IT2) Takagi-Sugeno (T-S) fuzzy identification, force feedback can be reconstructed at the master side without a sense of delay, thus ensuring the telepresence performance in the force-sensor-free scenarios. Two experimental applications validate the effectiveness of the proposed framework.

cs.RO↗

arXiv2021

Multiresolution Deep Implicit Functions for 3D Shape Representation

We introduce Multiresolution Deep Implicit Functions (MDIF), a hierarchical representation that can recover fine geometry detail, while being able to perform global operations such as shape completion. Our model represents a complex 3D shape with a hierarchy of latent grids, which can be decoded into different levels of detail and also achieve better accuracy. For shape completion, we propose latent grid dropout to simulate partial data in the latent space and therefore defer the completing functionality to the decoder side. This along with our multires design significantly improves the shape completion quality under decoder-only latent optimization. To the best of our knowledge, MDIF is the first deep implicit function model that can at the same time (1) represent different levels of detail and allow progressive decoding; (2) support both encoder-decoder inference and decoder-only latent optimization, and fulfill multiple applications; (3) perform detailed decoder-only shape completion. Experiments demonstrate its superior performance against prior art in various 3D reconstruction tasks.

cs.CV↗

arXiv2021

SofGAN: A Portrait Image Generator with Dynamic Styling

Recently, Generative Adversarial Networks (GANs)} have been widely used for portrait image generation. However, in the latent space learned by GANs, different attributes, such as pose, shape, and texture style, are generally entangled, making the explicit control of specific attributes difficult. To address this issue, we propose a SofGAN image generator to decouple the latent space of portraits into two subspaces: a geometry space and a texture space. The latent codes sampled from the two subspaces are fed to two network branches separately, one to generate the 3D geometry of portraits with canonical pose, and the other to generate textures. The aligned 3D geometries also come with semantic part segmentation, encoded as a semantic occupancy field (SOF). The SOF allows the rendering of consistent 2D semantic segmentation maps at arbitrary views, which are then fused with the generated texture maps and stylized to a portrait photo using our semantic instance-wise (SIW) module. Through extensive experiments, we show that our system can generate high quality portrait images with independently controllable geometry and texture attributes. The method also generalizes well in various applications such as appearance-consistent facial animation and dynamic styling.

cs.CV↗

arXiv2021

JUNO Physics and Detector

The Jiangmen Underground Neutrino Observatory (JUNO) is a 20 kton LS detector at 700-m underground. An excellent energy resolution and a large fiducial volume offer exciting opportunities for addressing many important topics in neutrino and astro-particle physics. With 6 years of data, the neutrino mass ordering can be determined at 3-4 sigma and three oscillation parameters can be measured to a precision of 0.6% or better by detecting reactor antineutrinos. With 10 years of data, DSNB could be observed at 3-sigma; a lower limit of the proton lifetime of 8.34e33 years (90% C.L.) can be set by searching for p->nu_bar K^+; detection of solar neutrinos would shed new light on the solar metallicity problem and examine the vacuum-matter transition region. A core-collapse supernova at 10 kpc would lead to ~5000 IBD and ~2000 (300) all-flavor neutrino-proton (electron) scattering events. Geo-neutrinos can be detected with a rate of ~400 events/year. We also summarize the final design of the JUNO detector and the key R&D achievements. All 20-inch PMTs have been tested. The average photon detection efficiency is 28.9% for the 15,000 MCP PMTs and 28.1% for the 5,000 dynode PMTs, higher than the JUNO requirement of 27%. Together with the >20 m attenuation length of LS, we expect a yield of 1345 p.e. per MeV and an effective energy resolution of 3.02%/\sqrt{E (MeV)}$ in simulations. The underwater electronics is designed to have a loss rate <0.5% in 6 years. With degassing membranes and a micro-bubble system, the radon concentration in the 35-kton water pool could be lowered to <10 mBq/m^3. Acrylic panels of radiopurity <0.5 ppt U/Th are produced. The 20-kton LS will be purified onsite. Singles in the fiducial volume can be controlled to ~10 Hz. The JUNO experiment also features a double calorimeter system with 25,600 3-inch PMTs, a LS testing facility OSIRIS, and a near detector TAO.

hep-ex↗

arXiv2021

The Design and Sensitivity of JUNO's scintillator radiopurity pre-detector OSIRIS

The OSIRIS detector is a subsystem of the liquid scintillator fillling chain of the JUNO reactor neutrino experiment. Its purpose is to validate the radiopurity of the scintillator to assure that all components of the JUNO scintillator system work to specifications and only neutrino-grade scintillator is filled into the JUNO Central Detector. The aspired sensitivity level of $10^{-16}$ g/g of $^{238}$U and $^{232}$Th requires a large ($\sim$20 m$^3$) detection volume and ultralow background levels. The present paper reports on the design and major components of the OSIRIS detector, the detector simulation as well as the measuring strategies foreseen and the sensitivity levels to U/Th that can be reached in this setup.

physics.ins-det↗

arXiv2021

Calibration Strategy of the JUNO Experiment

We present the calibration strategy for the 20 kton liquid scintillator central detector of the Jiangmen Underground Neutrino Observatory (JUNO). By utilizing a comprehensive multiple-source and multiple-positional calibration program, in combination with a novel dual calorimetry technique exploiting two independent photosensors and readout systems, we demonstrate that the JUNO central detector can achieve a better than 1% energy linearity and a 3% effective energy resolution, required by the neutrino mass ordering determination.

physics.ins-det↗

arXiv2020

Deep Eyes: Binocular Depth-from-Focus on Focal Stack Pairs

Human visual system relies on both binocular stereo cues and monocular focusness cues to gain effective 3D perception. In computer vision, the two problems are traditionally solved in separate tracks. In this paper, we present a unified learning-based technique that simultaneously uses both types of cues for depth inference. Specifically, we use a pair of focal stacks as input to emulate human perception. We first construct a comprehensive focal stack training dataset synthesized by depth-guided light field rendering. We then construct three individual networks: a Focus-Net to extract depth from a single focal stack, a EDoF-Net to obtain the extended depth of field (EDoF) image from the focal stack, and a Stereo-Net to conduct stereo matching. We show how to integrate them into a unified BDfF-Net to obtain high-quality depth maps. Comprehensive experiments show that our approach outperforms the state-of-the-art in both accuracy and speed and effectively emulates human vision systems.

cs.CV↗