arXiv subjects

Tao Wang

Publications and source records attributed to Tao Wang.

At least 433 records · Page 24Linked to original sources

arXiv2024

Feature Completion Transformer for Occluded Person Re-identification

Occluded person re-identification (Re-ID) is a challenging problem due to the destruction of occluders. Most existing methods focus on visible human body parts through some prior information. However, when complementary occlusions occur, features in occluded regions can interfere with matching, which affects performance severely. In this paper, different from most previous works that discard the occluded region, we propose a Feature Completion Transformer (FCFormer) to implicitly complement the semantic information of occluded parts in the feature space. Specifically, Occlusion Instance Augmentation (OIA) is proposed to simulates real and diverse occlusion situations on the holistic image. These augmented images not only enrich the amount of occlusion samples in the training set, but also form pairs with the holistic images. Subsequently, a dual-stream architecture with a shared encoder is proposed to learn paired discriminative features from pairs of inputs. Without additional semantic information, an occluded-holistic feature sample-label pair can be automatically created. Then, Feature Completion Decoder (FCD) is designed to complement the features of occluded regions by using learnable tokens to aggregate possible information from self-generated occluded features. Finally, we propose the Cross Hard Triplet (CHT) loss to further bridge the gap between complementing features and extracting features under the same ID. In addition, Feature Completion Consistency (FC$^2$) loss is introduced to help the generated completion feature distribution to be closer to the real holistic feature distribution. Extensive experiments over five challenging datasets demonstrate that the proposed FCFormer achieves superior performance and outperforms the state-of-the-art methods by significant margins on occluded datasets.

cs.CV↗

arXiv2024

Integrated distributed sensing and quantum communication networks

The integration of sensing and communication can achieve ubiquitous sensing while enabling ubiquitous communication. Within the gradually improving global communication, the integrated sensing and communication (ISAC) system based on optical fibers can accomplish various functionalities, such as urban structure imaging, seismic wave detection, and pipeline safety monitoring. With the development of quantum communication, quantum networks based on optical fiber are gradually being established. In this paper, we propose an integrated sensing and quantum network (ISAQN) scheme, which can achieve secure key distribution among multiple nodes and distributed sensing under the standard quantum limit. CV-QKD protocol and the round-trip multi-band structure are adopted to achieve the multi-node secure key distribution. Meanwhile, the spectrum phase monitoring (SPM) protocol is proposed to realize distributed sensing. It determines which node is vibrating by monitoring the frequency spectrum and restores the vibration waveform by monitoring the phase change. The scheme is experimentally demonstrated by simulating the vibration in a star structure network. Experimental results indicate that this multi-user quantum network can achieve a secret key rate (SKR) of approximately 0.7 $\rm{Mbits/s}$ for each user under 10 $\rm{km}$ standard fiber transmission and its network capacity is 8. In terms of distributed sensing, it can achieve a vibration response bandwidth ranging from 1 $\rm{Hz}$ to 2 $\rm{kHz}$, a strain resolution of 0.50 $\rm{n}$$\varepsilon$$/\sqrt{\rm{Hz}}$, and a spatial resolution of 0.20 $\rm{m}$ under shot-noise-limited detection. The proposed ISAQN scheme enables simultaneous quantum communication and distributed sensing in a multi-point network, laying a foundation for future large-scale quantum networks and high-precision sensing networks.

quant-ph↗

arXiv2024

Towards Real-World Blind Face Restoration with Generative Diffusion Prior

Blind face restoration is an important task in computer vision and has gained significant attention due to its wide-range applications. Previous works mainly exploit facial priors to restore face images and have demonstrated high-quality results. However, generating faithful facial details remains a challenging problem due to the limited prior knowledge obtained from finite data. In this work, we delve into the potential of leveraging the pretrained Stable Diffusion for blind face restoration. We propose BFRffusion which is thoughtfully designed to effectively extract features from low-quality face images and could restore realistic and faithful facial details with the generative prior of the pretrained Stable Diffusion. In addition, we build a privacy-preserving face dataset called PFHQ with balanced attributes like race, gender, and age. This dataset can serve as a viable alternative for training blind face restoration networks, effectively addressing privacy and bias concerns usually associated with the real face datasets. Through an extensive series of experiments, we demonstrate that our BFRffusion achieves state-of-the-art performance on both synthetic and real-world public testing datasets for blind face restoration and our PFHQ dataset is an available resource for training blind face restoration networks. The codes, pretrained models, and dataset are released at https://github.com/chenxx89/BFRffusion.

cs.CV↗

arXiv2024

Fewer-token Neural Speech Codec with Time-invariant Codes

Language model based text-to-speech (TTS) models, like VALL-E, have gained attention for their outstanding in-context learning capability in zero-shot scenarios. Neural speech codec is a critical component of these models, which can convert speech into discrete token representations. However, excessive token sequences from the codec may negatively affect prediction accuracy and restrict the progression of Language model based TTS models. To address this issue, this paper proposes a novel neural speech codec with time-invariant codes named TiCodec. By encoding and quantizing time-invariant information into a separate code, TiCodec can reduce the amount of frame-level information that needs encoding, effectively decreasing the number of tokens as codes of speech. Furthermore, this paper introduces a time-invariant encoding consistency loss to enhance the consistency of time-invariant code within an utterance and force it to capture more global information, which can benefit the zero-shot TTS task. Experimental results demonstrate that TiCodec can not only enhance the quality of reconstruction speech with fewer tokens but also increase the similarity and naturalness, as well as reduce the word error rate of the synthesized speech by the TTS model.

cs.SD↗

arXiv2024

GroundingGPT:Language Enhanced Multi-modal Grounding Model

Multi-modal large language models have demonstrated impressive performance across various tasks in different modalities. However, existing multi-modal models primarily emphasize capturing global information within each modality while neglecting the importance of perceiving local information across modalities. Consequently, these models lack the ability to effectively understand the fine-grained details of input data, limiting their performance in tasks that require a more nuanced understanding. To address this limitation, there is a compelling need to develop models that enable fine-grained understanding across multiple modalities, thereby enhancing their applicability to a wide range of tasks. In this paper, we propose GroundingGPT, a language enhanced multi-modal grounding model. Beyond capturing global information like other multi-modal models, our proposed model excels at tasks demanding a detailed understanding of local information within the input. It demonstrates precise identification and localization of specific regions in images or moments in videos. To achieve this objective, we design a diversified dataset construction pipeline, resulting in a multi-modal, multi-granularity dataset for model training. The code, dataset, and demo of our model can be found at https: //github.com/lzw-lzw/GroundingGPT.

cs.CV↗

arXiv2024

Zero-Shot Aerial Object Detection with Visual Description Regularization

Existing object detection models are mainly trained on large-scale labeled datasets. However, annotating data for novel aerial object classes is expensive since it is time-consuming and may require expert knowledge. Thus, it is desirable to study label-efficient object detection methods on aerial images. In this work, we propose a zero-shot method for aerial object detection named visual Description Regularization, or DescReg. Concretely, we identify the weak semantic-visual correlation of the aerial objects and aim to address the challenge with prior descriptions of their visual appearance. Instead of directly encoding the descriptions into class embedding space which suffers from the representation gap problem, we propose to infuse the prior inter-class visual similarity conveyed in the descriptions into the embedding learning. The infusion process is accomplished with a newly designed similarity-aware triplet loss which incorporates structured regularization on the representation space. We conduct extensive experiments with three challenging aerial object detection datasets, including DIOR, xView, and DOTA. The results demonstrate that DescReg significantly outperforms the state-of-the-art ZSD methods with complex projection designs and generative frameworks, e.g., DescReg outperforms best reported ZSD method on DIOR by 4.5 mAP on unseen classes and 8.1 in HM. We further show the generalizability of DescReg by integrating it into generative ZSD methods as well as varying the detection architecture.

cs.CV↗

arXiv2024

Maximum odd induced subgraph of a graph concerning its chromatic number

Let $f_{o}(G)$ be the maximum order of an odd induced subgraph of $G$. In 1992, Scott proposed a conjecture that $f_{o}(G)\geq \frac {n} {2χ(G)}$ for a graph $G$ of order $n$ without isolated vertices, where $χ(G)$ is the chromatic number of $G$. In this paper, we show that the conjecture is not true for bipartite graphs, but is true for all line graphs. In addition, we also disprove a conjecture of Berman, Wang and Wargo in 1997, which states that $f_{o}(G)\geq 2\lfloor\frac {n} {4}\rfloor$ for a connected graph $G$ of order $n$. Scott's conjecture is open for a graph with chromatic number at least 3.

math.CO↗

arXiv2024

Comparison of sectoral structures between China and Japan: A network perspective

Economic structure comparisons between China and Japan have long captivated development economists. To delve deeper into their sectoral differences from 1995 to 2018, we used the annual input-output tables (IOTs) of both nations to construct weighted and directed input-output networks (IONs). This facilitated deeper network analyses. Strength distributions underscored variations in inter-sector economic interactions. Weighted, directed assortativity coefficients encapsulated the homophily among connecting sectors' features. By adjusting emphasis in PageRank centrality, key sectors were identified. Community detection revealed their clustering tendencies among the sectors. As anticipated, the analysis pinpointed manufacturing as China's central sector, while Japan favored services. Yet, at a finer level of the specific sectors, both nations exhibited varied structural evolutions. Contrastingly, sectoral communities in both China and Japan demonstrated commendable stability over the examined duration.

stat.AP↗

arXiv2024

Symmetry and functional inequalities for stable Lévy-type operators

In this paper, we provide the sufficient and necessary conditions for the symmetry of the following stable Lévy-type operator $\mathcal{L}$ on $\mathbb{R}$: $$\mathcal{L}=a(x){Δ^{α/2}}+b(x)\frac{\d}{\d x},$$ where $a,b$ are the continuous positive and differentiable functions, respectively. Under the assumption of symmetry, we further study the criteria for functional inequalities, including Poincaré inequalities, logarithmic Sobolev inequalities and Nash inequalities. Our proofs rely on the Orlicz space theory and the estimates of the Green functions.

math.PR↗

arXiv2024

Cosmic evolution of radio-excess AGNs in quiescent and star-forming galaxies across $0 < z < 4$

Recent deep and wide radio surveys extend the studies for radio-excess active galactic nuclei (radio-AGNs) to lower luminosities and higher redshifts, providing new insights into the abundance and physical origin of radio-AGNs. Here we focus on the cosmic evolution, physical properties and AGN-host galaxy connections of radio-AGNs selected from a sample of ~ 500,000 galaxies at 0 < z < 4 in GOODS-N, GOODS-S, and COSMOS fields. Combining deep radio data with multi-band, de-blended far-infrared (FIR) and sub-millimeter data, we identify 1162 radio-AGNs through radio excess relative to the FIR-radio relation. We study the cosmic evolution of 1.4 GHz radio luminosity functions (RLFs) for star-forming galaxies (SFGs) and radio-AGNs, which are well described by a pure luminosity evolution of $L_*\propto (1+z)^{-0.31z+3.41}$ and a pure density evolution of $Φ_*\propto (1+z)^{-0.80z+2.88}$, respectively. We derive the turnover luminosity above which the number density of radio-AGNs surpasses that of SFGs. This crossover luminosity increases as increasing redshift, from $10^{22.9}$ W Hz$^{-1}$ at z ~ 0 to $10^{25.2}$ W Hz$^{-1}$ at z ~ 4. At full redshift range (0 < z < 4), we further derive the probability ($p_{radio}$) of SFGs and quiescent galaxies (QGs) hosting a radio-AGN as a function of stellar mass ($M_*$), radio luminosity ($L_R$), and redshift (z), which yields $p_{radio}\propto (1+z)^{3.54}M_*^{1.02}L_R^{-0.90}$ for SFGs, and $p_{radio}\propto (1+z)^{2.38}M_*^{1.39}L_R^{-0.60}$ for QGs, respectively. It indicates that radio-AGNs in QGs prefer to reside in more massive galaxies with larger $L_R$ than those in SFGs, and radio-AGN fraction increases towards higher redshift in both SFGs and QGs with a more rapid increase in SFGs. Further, we find that the radio-AGN fraction depends on accretion states of BHs and redshift in SFGs, while in QGs it also depends on BH (or galaxy) mass.

astro-ph.GA↗

arXiv2024

A version of Bakry-Émery Ricci flow on a finite graph

In this paper, we study the Bakry-Émery Ricci flow on finite graphs. Our main result is the local existence and uniqueness of solutions to the Ricci flow. We prove the long-time convergence or finite-time blow up for the Bakry-Émery Ricci flow on finite trees and circles.

math.DG↗

arXiv2024

Sensor Misalignment-tolerant AUV Navigation with Passive DoA and Doppler Measurements

We present a sensor misalignment-tolerant AUV navigation method that leverages measurements from an acoustic array and dead reckoned information. Recent studies have demonstrated the potential use of passive acoustic Direction of Arrival (DoA) measurements for AUV navigation without requiring ranging measurements. However, the sensor misalignment between the acoustic array and the attitude sensor was not accounted for. Such misalignment may deteriorate the navigation accuracy. This paper proposes a novel approach that allows simultaneous AUV navigation, beacon localization, and sensor alignment. An Unscented Kalman Filter (UKF) that enables the necessary calculations to be completed at an affordable computational load is developed. A Nonlinear Least Squares (NLS)-based technique is employed to find an initial solution for beacon localization and sensor alignment as early as possible using a short-term window of measurements. Experimental results demonstrate the performance of the proposed method.

cs.RO↗

arXiv2024

Anything in Any Scene: Photorealistic Video Object Insertion

Realistic video simulation has shown significant potential across diverse applications, from virtual reality to film production. This is particularly true for scenarios where capturing videos in real-world settings is either impractical or expensive. Existing approaches in video simulation often fail to accurately model the lighting environment, represent the object geometry, or achieve high levels of photorealism. In this paper, we propose Anything in Any Scene, a novel and generic framework for realistic video simulation that seamlessly inserts any object into an existing dynamic video with a strong emphasis on physical realism. Our proposed general framework encompasses three key processes: 1) integrating a realistic object into a given scene video with proper placement to ensure geometric realism; 2) estimating the sky and environmental lighting distribution and simulating realistic shadows to enhance the light realism; 3) employing a style transfer network that refines the final video output to maximize photorealism. We experimentally demonstrate that Anything in Any Scene framework produces simulated videos of great geometric realism, lighting realism, and photorealism. By significantly mitigating the challenges associated with video data generation, our framework offers an efficient and cost-effective solution for acquiring high-quality videos. Furthermore, its applications extend well beyond video data augmentation, showing promising potential in virtual reality, video editing, and various other video-centric applications. Please check our project website https://anythinginanyscene.github.io for access to our project code and more high-resolution video results.

cs.CV↗

arXiv2024

Dual-Decoder Consistency via Pseudo-Labels Guided Data Augmentation for Semi-Supervised Medical Image Segmentation

While supervised learning has achieved remarkable success, obtaining large-scale labeled datasets in biomedical imaging is often impractical due to high costs and the time-consuming annotations required from radiologists. Semi-supervised learning emerges as an effective strategy to overcome this limitation by leveraging useful information from unlabeled datasets. In this paper, we present a novel semi-supervised learning method, Dual-Decoder Consistency via Pseudo-Labels Guided Data Augmentation (DCPA), for medical image segmentation. We devise a consistency regularization to promote consistent representations during the training process. Specifically, we use distinct decoders for student and teacher networks while maintain the same encoder. Moreover, to learn from unlabeled data, we create pseudo-labels generated by the teacher networks and augment the training data with the pseudo-labels. Both techniques contribute to enhancing the performance of the proposed method. The method is evaluated on three representative medical image segmentation datasets. Comprehensive comparisons with state-of-the-art semi-supervised medical image segmentation methods were conducted under typical scenarios, utilizing 10% and 20% labeled data, as well as in the extreme scenario of only 5% labeled data. The experimental results consistently demonstrate the superior performance of our method compared to other methods across the three semi-supervised settings. The source code is publicly available at https://github.com/BinYCn/DCPA.git.

eess.IV↗

arXiv2024

Non-integer Floquet Sidebands Spectroscopy

In the quantum system under periodical modulation, the particle can be excited by absorbing the laser photon with the assistance of integer Floquet photons, so that the Floquet sidebands appear. Here, we experimentally observe non-integer Floquet sidebands (NIFBs) emerging between the integer ones while increasing the strength of the probe laser in the optical lattice clock system. Then, we propose the Floquet channel interference hypothesis (FCIH) which surprisingly matches quantitatively well with both experimental and numerical results. With its help, we found both Rabi and Ramsey spectra are very sensitive to the initial phase and exhibit additional two symmetries. More importantly, the height of Ramsey NIFBs is comparable to the integer one at larger $g/ω_s$ which indicates an exotic phenomenon beyond the perturbative description. Our work provides new insight into the spectroscopy of the Floquet system and has potential application in quantum technology.

cond-mat.quant-gas↗

arXiv2024

Understanding CNNs from excitations

Saliency maps have proven to be a highly efficacious approach for explicating the decisions of Convolutional Neural Networks. However, extant methodologies predominantly rely on gradients, which constrain their ability to explicate complex models. Furthermore, such approaches are not fully adept at leveraging negative gradient information to improve interpretive veracity. In this study, we present a novel concept, termed positive and negative excitation, which enables the direct extraction of positive and negative excitation for each layer, thus enabling complete layer-by-layer information utilization sans gradients. To organize these excitations into final saliency maps, we introduce a double-chain backpropagation procedure. A comprehensive experimental evaluation, encompassing both binary classification and multi-classification tasks, was conducted to gauge the effectiveness of our proposed method. Encouragingly, the results evince that our approach offers a significant improvement over the state-of-the-art methods in terms of salient pixel removal, minor pixel removal, and inconspicuous adversarial perturbation generation guidance. Additionally, we verify the correlation between positive and negative excitations.

cs.CV↗

arXiv2024

Ultrasensitive piezoelectric sensor based on two-dimensional Na2Cl crystals with periodic atom vacancies

Pursuing ultrasensitivity of pressure sensors has been a long-standing goal. Here, we report a piezoelectric sensor that exhibits supreme pressure-sensing performance, including a peak sensitivity up to 3.5*10^6 kPa^-1 in the pressure range of 1-100 mPa and a detection limit of less than 1 mPa, superior to the current state-of-the-art pressure sensors. These properties are attributed to the high percentage of periodic atom vacancies in the two-dimensional Na2Cl crystals formed within multilayered graphene oxide membrane in the sensor, which provides giant polarization with high stability. The sensor can even clearly detect the airflow fluctuations surrounding a flapping butterfly, which have long been the elusive tiny signals in the famous "butterfly effect". The finding represents a step towards next-generation pressure sensors for various precision applications.

physics.app-ph↗

arXiv2024

Individual subpulses of PSR B1916+14 and their polarization properties

Individual subpulses of pulsars are regarded as the basic emission components, providing invaluable information to understand the radio emission process in the pulsar magnetosphere. Nevertheless, subpulses are overlapped with each other along the rotation phase for most pulsars, making it difficult to study the statistical properties of subpulses. Among the pulsars observed by the Five-hundred-meter Aperture Spherical radio Telescope, PSR B1916+14 has a large number of isolated well-resolved subpulses in the high time resolution observations, having a typical width of 0.15 ms and a high linear polarization. We find that the number distribution of subpulses contributes dominantly to the mean profile. According to the emission geometry, these emission units come from a region roughly 155 km above the polar cap in the pulsar magnetosphere, and the length scale of basic emission units is approximately 120 m. The deviations of polarization position angles for these single subpulses from the standard S-shaped curve are closely related to their fractional linear and circular polarization, and the large deviations tend to come from drifting subpulses.

astro-ph.HE↗