arXiv subjects

Jie Song

Publications and source records attributed to Jie Song.

At least 163 records · Page 9Linked to original sources

arXiv2023

Thermal-dephasing-tolerant generation of Schrödinger cat states with Rydberg dressed blockade

Multipartite entangled states involving non-locality are one of the most fascinating characteristics of quantum mechanics. In this work, we propose a thermal-dephasing-tolerant generation of mesoscopic entangled states with Rydberg dressed atoms. We encode logical state on dressed states rather than Rydberg states. Such treatment can increase the lifetime of multipartite entanglement coherence to around 3 times compared to the Rydberg-state-coding one at the same system size, and therefore induce solid fidelities of mesoscopic superposition states generation. The current work theoretically verifies the advantages of using Rydberg dressed states in many-body quantum entanglement, which is helpful for large-scale quantum computation and many-body Rydberg quantum simulation.

quant-ph↗

arXiv2023

Human from Blur: Human Pose Tracking from Blurry Images

We propose a method to estimate 3D human poses from substantially blurred images. The key idea is to tackle the inverse problem of image deblurring by modeling the forward problem with a 3D human model, a texture map, and a sequence of poses to describe human motion. The blurring process is then modeled by a temporal image aggregation step. Using a differentiable renderer, we can solve the inverse problem by backpropagating the pixel-wise reprojection error to recover the best human motion representation that explains a single or multiple input images. Since the image reconstruction loss alone is insufficient, we present additional regularization terms. To the best of our knowledge, we present the first method to tackle this problem. Our method consistently outperforms other methods on significantly blurry inputs since they lack one or multiple key functionalities that our method unifies, i.e. image deblurring with sub-frame accuracy and explicit 3D modeling of non-rigid human motion.

cs.CV↗

arXiv2023

ModelGiF: Gradient Fields for Model Functional Distance

The last decade has witnessed the success of deep learning and the surge of publicly released trained models, which necessitates the quantification of the model functional distance for various purposes. However, quantifying the model functional distance is always challenging due to the opacity in inner workings and the heterogeneity in architectures or tasks. Inspired by the concept of "field" in physics, in this work we introduce Model Gradient Field (abbr. ModelGiF) to extract homogeneous representations from the heterogeneous pre-trained models. Our main assumption underlying ModelGiF is that each pre-trained deep model uniquely determines a ModelGiF over the input space. The distance between models can thus be measured by the similarity between their ModelGiFs. We validate the effectiveness of the proposed ModelGiF with a suite of testbeds, including task relatedness estimation, intellectual property protection, and model unlearning verification. Experimental results demonstrate the versatility of the proposed ModelGiF on these tasks, with significantly superiority performance to state-of-the-art competitors. Codes are available at https://github.com/zju-vipa/modelgif.

cs.LG↗

arXiv2023

EMDB: The Electromagnetic Database of Global 3D Human Pose and Shape in the Wild

We present EMDB, the Electromagnetic Database of Global 3D Human Pose and Shape in the Wild. EMDB is a novel dataset that contains high-quality 3D SMPL pose and shape parameters with global body and camera trajectories for in-the-wild videos. We use body-worn, wireless electromagnetic (EM) sensors and a hand-held iPhone to record a total of 58 minutes of motion data, distributed over 81 indoor and outdoor sequences and 10 participants. Together with accurate body poses and shapes, we also provide global camera poses and body root trajectories. To construct EMDB, we propose a multi-stage optimization procedure, which first fits SMPL to the 6-DoF EM measurements and then refines the poses via image observations. To achieve high-quality results, we leverage a neural implicit avatar model to reconstruct detailed human surface geometry and appearance, which allows for improved alignment and smoothness via a dense pixel-level objective. Our evaluations, conducted with a multi-view volumetric capture system, indicate that EMDB has an expected accuracy of 2.3 cm positional and 10.6 degrees angular error, surpassing the accuracy of previous in-the-wild datasets. We evaluate existing state-of-the-art monocular RGB methods for camera-relative and global pose estimation on EMDB. EMDB is publicly available under https://ait.ethz.ch/emdb

cs.CV↗

arXiv2023

Strong [O III] λ5007 Compact Galaxies Identified from SDSS DR16 and Their Scaling Relations

Green pea galaxies are a special class of star-forming compact galaxies with strong [O III]λ5007 and considered as analogs of high-redshift Lyα-emitting galaxies and potential sources for cosmic reionization. In this paper, we identify 76 strong [O III]λ5007 compact galaxies at z < 0.35 from DR1613 of the Sloan Digital Sky Survey. These galaxies present relatively low stellar mass, high star formation rate, and low metallicity. Both star-forming main sequence relation (SFMS) and mass-metallicity relation (MZR) are investigated and compared with green pea and blueberry galaxies collected from literature. It is found that our strong [O III] λ5007 compact galaxies share common properties with those compact galaxies with extreme star formation and show distinct scaling relations in respect to those of normal star-forming galaxies at the same redshift. The slope of SFMS is higher, indicates that strong [O III]λ5007 compact galaxies might grow faster in stellar mass. The lower MZR implies that they may be less chemically evolved and hence on the early stage of star formation. A further environmental investigation confirms that they inhabit relatively low-density regions. Future largescale spectroscopic surveys will provide more details on their physical origin and evolution.

astro-ph.GA↗

arXiv2023

Evolution of Non-parametric Morphology of Galaxies in the JWST CEERS Field at $z\simeq$0.8-3.0

Galaxy morphology is one of the most fundamental ways to describe galaxy properties, but the morphology we observe may be affected by wavelength and spatial resolution, which may introduce systematic bias when comparing galaxies at different redshift. Taking advantage of the broad wavelength coverage from optical to near-IR and high resolution NIRCam instrument of JWST, we measure the non-parametric morphological parameters of a total of 1376 galaxies at $z\simeq$0.8-3.0 in the CEERS field through an optimized code called {\tt\string statmorph\_csst}. We divide our sample into three redshift intervals and investigate the wavelength- and redshift-dependence of the morphological parameters. We also explore how the widely-used galaxy type classification methods based on the morphological parameters depend on wavelength and spatial resolution. We find that there are variations in all morphological parameters with rest-frame wavelength ($λ_{\rm rf}$), especially at the short wavelength end, and the $λ_{\rm rf}$ mainly affects the classification between late-type and early-type galaxy. As the $λ_{\rm rf}$ increases, the galaxies on the $G-M_{20}$ diagram move to the upper left with a slope of -0.23$\pm$0.03 on average. We find that spatial resolution mainly affects the merger identification. The merger fraction in F200W resolution can be $\ga$2 times larger than that in F444W resolution. Furthermore, We compare the morphological parameter evolution of galaxies with different stellar masses. We find that there are differences in the morphological evolution of high- and low-mass (log$M_*\geqslant$10 and 9$<$log$M_*<$10) galaxies in the studied redshift range, which may be caused by their different evolution paths.

astro-ph.GA↗

arXiv2023

Quantum parametric amplifiation of phonon-mediated magnon-spin interaction

The recently developed hybrid magnonics provides new opportunities for advances in both the study of magnetism and the development of quantum information processing. However, engineering coherent quantum state transfer between magnons and specific information carriers, in particular, mechanical oscillators and solid-state spins, remains challenging due to the intrinsically weak interactions and the frequency mismatch between diffrent components. Here, we show how to strongly couple the magnon modes in a nanomagnet to the quantized mechanical motion (phonons) of a micromechanical cantilever in a hybrid tripartite system. The coherent and enhanced magnon-phonon coupling is engineered by introducing the quantum parametric amplifiation of the mechanical motion. With experimentally feasible parameters, we show that the mechanical parametric drive can be adjusted to drive the system into the strong-coupling regime and even the ultrastrong-coupling regime. Furthermore, we show the coherent state transfer between the nanomagnet and a nitrogen-vacancy center in the dispersive-coupling regime, with the magnon-spin interaction mediated by the virtually-excited squeezed phonons. The amplifid mechanical noise can hardly interrupt the coherent dynamics of the system even for low mechanical quality factors, which removes the requirement of applying additional engineered-reservoir techniques. Our work opens up prospects for developing novel quantum transducers, quantum memories and high-precision measurements.

quant-ph↗

arXiv2023

Schema Inference for Interpretable Image Classification

In this paper, we study a novel inference paradigm, termed as schema inference, that learns to deductively infer the explainable predictions by rebuilding the prior deep neural network (DNN) forwarding scheme, guided by the prevalent philosophical cognitive concept of schema. We strive to reformulate the conventional model inference pipeline into a graph matching policy that associates the extracted visual concepts of an image with the pre-computed scene impression, by analogy with human reasoning mechanism via impression matching. To this end, we devise an elaborated architecture, termed as SchemaNet, as a dedicated instantiation of the proposed schema inference concept, that models both the visual semantics of input instances and the learned abstract imaginations of target categories as topological relational graphs. Meanwhile, to capture and leverage the compositional contributions of visual semantics in a global view, we also introduce a universal Feat2Graph scheme in SchemaNet to establish the relational graphs that contain abundant interaction information. Both the theoretical analysis and the experimental results on several benchmarks demonstrate that the proposed schema inference achieves encouraging performance and meanwhile yields a clear picture of the deductive process leading to the predictions. Our code is available at https://github.com/zhfeing/SchemaNet-PyTorch.

cs.CV↗

arXiv2023

Robust transitionless quantum driving: Concatenated approach

We propose a concatenated approach for implementing transitionless quantum driving regardless of adiabatic conditions while being robustness with respect to all kinds of systematic errors induced by pulse duration, pulse amplitude, detunings, and Stark shift, etc. The current approach is particularly efficient for all time-dependent pulses with arbitrary shape, and only the phase differences between pulses is required to properly modulate. The simple physical implementation without the help of pulse shaping techniques or extra pulses makes this approach quite universal and provides a different avenue for robust quantum control by the time-dependent Hamiltonian.

quant-ph↗

arXiv2023

The Classification of Galaxy Morphology in H-band of COSMOS-DASH Field: a combination-based machine learning clustering model

By applying our previously developed two-step scheme for galaxy morphology classification, we present a catalog of galaxy morphology for H-band selected massive galaxies in the COSMOS-DASH field, which includes 17292 galaxies with stellar mass $M_{\star}>10^{10}~M_{\odot}$ at $0.5<z<2.5$. The classification scheme is designed to provide a complete morphology classification for galaxies via a combination of two machine-learning steps. We first use an unsupervised machine learning method (i.e., bagging-based multi-clustering) to cluster galaxies into five categories: spherical (SPH), early-type disk (ETD), late-type disk (LTD), irregular (IRR), and unclassified (UNC). About 48\% of galaxies (8258/17292) are successfully clustered during this step. For the remaining sample, we adopt a supervised machine learning method (i.e., GoogLeNet) to classify them, during which galaxies that are well-classified in the previous step are taken as our training set. Consequently, we obtain a morphology classification result for the full sample. The t-SNE test shows that galaxies in our sample can be well aggregated. We also measure the parametric and nonparametric morphologies of these galaxies. We find that the Sérsic index increases from IRR to SPH and the effective radius decreases from IRR to SPH, consistent with the corresponding definitions. Galaxies from different categories are separately distributed in the $G$--$M_{20}$ space. Such consistencies with other characteristic descriptions of galaxy morphology demonstrate the reliability of our classification result, ensuring that it can be used as a basic catalog for further galaxy studies.

astro-ph.GA↗

arXiv2023

The effect of environment on the properties of the most massive galaxies at $0.5<z<2.5$ in the cosmos-dash field

How the environment influences the most massive galaxies is still unclear. To explore the environmental effects on morphology and star formation in the most massive galaxies at high redshift, we select galaxies with stellar mass $\log(M_{\star}/M_{\odot})>11$ at $0.5<z<2.5$ in the COSMOS-DASH field, which is the largest field with near-infrared photometrical observations using HST/WFC3 to date. Combining with the newly published COSMOS2020 catalog, we estimate the localized galaxy overdensity using a density estimator within the Bayesian probability framework. With the overdensity map, no significant environmental dependence is found in the distributions of Sérsic index and effective radius. When we consider the star formation state, galaxies in lower density are found to have higher median specific star formation rate (sSFR) at $0.5<z<1.5$. But for star-forming galaxies only, sSFR is independent of the environment within the whole redshift range, indicating that the primary effect of the environment might be to control the quiescent fraction. Based on these observations, the possible environmental quenching process for these massive galaxies might be mergers.

astro-ph.GA↗

arXiv2023

Learning Locally Editable Virtual Humans

In this paper, we propose a novel hybrid representation and end-to-end trainable network architecture to model fully editable and customizable neural avatars. At the core of our work lies a representation that combines the modeling power of neural fields with the ease of use and inherent 3D consistency of skinned meshes. To this end, we construct a trainable feature codebook to store local geometry and texture features on the vertices of a deformable body model, thus exploiting its consistent topology under articulation. This representation is then employed in a generative auto-decoder architecture that admits fitting to unseen scans and sampling of realistic avatars with varied appearances and geometries. Furthermore, our representation allows local editing by swapping local features between 3D assets. To verify our method for avatar creation and editing, we contribute a new high-quality dataset, dubbed CustomHumans, for training and evaluation. Our experiments quantitatively and qualitatively show that our method generates diverse detailed avatars and achieves better model fitting performance compared to state-of-the-art methods. Our code and dataset are available at https://custom-humans.github.io/.

cs.CV↗

arXiv2023

Temporal Aggregation and Propagation Graph Neural Networks for Dynamic Representation

Temporal graphs exhibit dynamic interactions between nodes over continuous time, whose topologies evolve with time elapsing. The whole temporal neighborhood of nodes reveals the varying preferences of nodes. However, previous works usually generate dynamic representation with limited neighbors for simplicity, which results in both inferior performance and high latency of online inference. Therefore, in this paper, we propose a novel method of temporal graph convolution with the whole neighborhood, namely Temporal Aggregation and Propagation Graph Neural Networks (TAP-GNN). Specifically, we firstly analyze the computational complexity of the dynamic representation problem by unfolding the temporal graph in a message-passing paradigm. The expensive complexity motivates us to design the AP (aggregation and propagation) block, which significantly reduces the repeated computation of historical neighbors. The final TAP-GNN supports online inference in the graph stream scenario, which incorporates the temporal information into node embeddings with a temporal activation function and a projection layer besides several AP blocks. Experimental results on various real-life temporal networks show that our proposed TAP-GNN outperforms existing temporal graph methods by a large margin in terms of both predictive performance and online inference latency. Our code is available at \url{https://github.com/doujiang-zheng/TAP-GNN}.

cs.LG↗

arXiv2023

Robust beam splitter with fast quantum state transfer through a topological interface

The Su-Schrieffer-Heeger (SSH) model, commonly used for robust state transfers through topologically protected edge pumping, has been generalized and exploited to engineer diverse functional quantum devices. Here, we propose to realize a fast topological beam splitter based on a generalized SSH model by accelerating the quantum state transfer (QST) process essentially limited by adiabatic requirements. The scheme involves delicate orchestration of the instantaneous energy spectrum through exponential modulation of nearest neighbor coupling strengths and onsite energies, yielding a significantly accelerated beam splitting process. Due to properties of topological pumping and accelerated QST, the beam splitter exhibits strong robustness against parameter disorders and losses of system. In addition, the model demonstrates good scalability and can be extended to two-dimensional crossed-chain structures to realize a topological router with variable numbers of output ports. Our work provides practical prospects for fast and robust topological QST in feasible quantum devices in large-scale quantum information processing.

quant-ph↗

arXiv2023

Hi4D: 4D Instance Segmentation of Close Human Interaction

We propose Hi4D, a method and dataset for the automatic analysis of physically close human-human interaction under prolonged contact. Robustly disentangling several in-contact subjects is a challenging task due to occlusions and complex shapes. Hence, existing multi-view systems typically fuse 3D surfaces of close subjects into a single, connected mesh. To address this issue we leverage i) individually fitted neural implicit avatars; ii) an alternating optimization scheme that refines pose and surface through periods of close proximity; and iii) thus segment the fused raw scans into individual instances. From these instances we compile Hi4D dataset of 4D textured scans of 20 subject pairs, 100 sequences, and a total of more than 11K frames. Hi4D contains rich interaction-centric annotations in 2D and 3D alongside accurately registered parametric body models. We define varied human pose and shape estimation tasks on this dataset and provide results from state-of-the-art methods on these benchmarks.

cs.CV↗

arXiv2023

Generalization Matters: Loss Minima Flattening via Parameter Hybridization for Efficient Online Knowledge Distillation

Most existing online knowledge distillation(OKD) techniques typically require sophisticated modules to produce diverse knowledge for improving students' generalization ability. In this paper, we strive to fully utilize multi-model settings instead of well-designed modules to achieve a distillation effect with excellent generalization performance. Generally, model generalization can be reflected in the flatness of the loss landscape. Since averaging parameters of multiple models can find flatter minima, we are inspired to extend the process to the sampled convex combinations of multi-student models in OKD. Specifically, by linearly weighting students' parameters in each training batch, we construct a Hybrid-Weight Model(HWM) to represent the parameters surrounding involved students. The supervision loss of HWM can estimate the landscape's curvature of the whole region around students to measure the generalization explicitly. Hence we integrate HWM's loss into students' training and propose a novel OKD framework via parameter hybridization(OKDPH) to promote flatter minima and obtain robust solutions. Considering the redundancy of parameters could lead to the collapse of HWM, we further introduce a fusion operation to keep the high similarity of students. Compared to the state-of-the-art(SOTA) OKD methods and SOTA methods of seeking flat minima, our OKDPH achieves higher performance with fewer parameters, benefiting OKD with lightweight and robust characteristics. Our code is publicly available at https://github.com/tianlizhang/OKDPH.

cs.CV↗

arXiv2023

Contrastive Identity-Aware Learning for Multi-Agent Value Decomposition

Value Decomposition (VD) aims to deduce the contributions of agents for decentralized policies in the presence of only global rewards, and has recently emerged as a powerful credit assignment paradigm for tackling cooperative Multi-Agent Reinforcement Learning (MARL) problems. One of the main challenges in VD is to promote diverse behaviors among agents, while existing methods directly encourage the diversity of learned agent networks with various strategies. However, we argue that these dedicated designs for agent networks are still limited by the indistinguishable VD network, leading to homogeneous agent behaviors and thus downgrading the cooperation capability. In this paper, we propose a novel Contrastive Identity-Aware learning (CIA) method, explicitly boosting the credit-level distinguishability of the VD network to break the bottleneck of multi-agent diversity. Specifically, our approach leverages contrastive learning to maximize the mutual information between the temporal credits and identity representations of different agents, encouraging the full expressiveness of credit assignment and further the emergence of individualities. The algorithm implementation of the proposed CIA module is simple yet effective that can be readily incorporated into various VD architectures. Experiments on the SMAC benchmarks and across different VD backbones demonstrate that the proposed method yields results superior to the state-of-the-art counterparts. Our code is available at https://github.com/liushunyu/CIA.

cs.LG↗

arXiv2023

X-Avatar: Expressive Human Avatars

We present X-Avatar, a novel avatar model that captures the full expressiveness of digital humans to bring about life-like experiences in telepresence, AR/VR and beyond. Our method models bodies, hands, facial expressions and appearance in a holistic fashion and can be learned from either full 3D scans or RGB-D data. To achieve this, we propose a part-aware learned forward skinning module that can be driven by the parameter space of SMPL-X, allowing for expressive animation of X-Avatars. To efficiently learn the neural shape and deformation fields, we propose novel part-aware sampling and initialization strategies. This leads to higher fidelity results, especially for smaller body parts while maintaining efficient training despite increased number of articulated bones. To capture the appearance of the avatar with high-frequency details, we extend the geometry and deformation fields with a texture network that is conditioned on pose, facial expression, geometry and the normals of the deformed surface. We show experimentally that our method outperforms strong baselines in both data domains both quantitatively and qualitatively on the animation task. To facilitate future research on expressive avatars we contribute a new dataset, called X-Humans, containing 233 sequences of high-quality textured scans from 20 participants, totalling 35,500 data frames.

cs.CV↗