arXiv subjects

Bo Zhang

Publications and source records attributed to Bo Zhang.

At least 487 records · Page 27Linked to original sources

arXiv2023

LogLG: Weakly Supervised Log Anomaly Detection via Log-Event Graph Construction

Fully supervised log anomaly detection methods suffer the heavy burden of annotating massive unlabeled log data. Recently, many semi-supervised methods have been proposed to reduce annotation costs with the help of parsed templates. However, these methods consider each keyword independently, which disregards the correlation between keywords and the contextual relationships among log sequences. In this paper, we propose a novel weakly supervised log anomaly detection framework, named LogLG, to explore the semantic connections among keywords from sequences. Specifically, we design an end-to-end iterative process, where the keywords of unlabeled logs are first extracted to construct a log-event graph. Then, we build a subgraph annotator to generate pseudo labels for unlabeled log sequences. To ameliorate the annotation quality, we adopt a self-supervised task to pre-train a subgraph annotator. After that, a detection model is trained with the generated pseudo labels. Conditioned on the classification results, we re-extract the keywords from the log sequences and update the log-event graph for the next iteration. Experiments on five benchmarks validate the effectiveness of LogLG for detecting anomalies on unlabeled log data and demonstrate that LogLG, as the state-of-the-art weakly supervised method, achieves significant performance improvements compared to existing methods.

cs.SE↗

arXiv2023

Make-It-3D: High-Fidelity 3D Creation from A Single Image with Diffusion Prior

In this work, we investigate the problem of creating high-fidelity 3D content from only a single image. This is inherently challenging: it essentially involves estimating the underlying 3D geometry while simultaneously hallucinating unseen textures. To address this challenge, we leverage prior knowledge from a well-trained 2D diffusion model to act as 3D-aware supervision for 3D creation. Our approach, Make-It-3D, employs a two-stage optimization pipeline: the first stage optimizes a neural radiance field by incorporating constraints from the reference image at the frontal view and diffusion prior at novel views; the second stage transforms the coarse model into textured point clouds and further elevates the realism with diffusion prior while leveraging the high-quality textures from the reference image. Extensive experiments demonstrate that our method outperforms prior works by a large margin, resulting in faithful reconstructions and impressive visual quality. Our method presents the first attempt to achieve high-quality 3D creation from a single image for general objects and enables various applications such as text-to-3D creation and texture editing.

cs.CV↗

arXiv2023

Generative Diffusion Prior for Unified Image Restoration and Enhancement

Existing image restoration methods mostly leverage the posterior distribution of natural images. However, they often assume known degradation and also require supervised training, which restricts their adaptation to complex real applications. In this work, we propose the Generative Diffusion Prior (GDP) to effectively model the posterior distributions in an unsupervised sampling manner. GDP utilizes a pre-train denoising diffusion generative model (DDPM) for solving linear inverse, non-linear, or blind problems. Specifically, GDP systematically explores a protocol of conditional guidance, which is verified more practical than the commonly used guidance way. Furthermore, GDP is strength at optimizing the parameters of degradation model during the denoising process, achieving blind image restoration. Besides, we devise hierarchical guidance and patch-based methods, enabling the GDP to generate images of arbitrary resolutions. Experimentally, we demonstrate GDP's versatility on several image datasets for linear problems, such as super-resolution, deblurring, inpainting, and colorization, as well as non-linear and blind issues, such as low-light enhancement and HDR image recovery. GDP outperforms the current leading unsupervised methods on the diverse benchmarks in reconstruction quality and perceptual quality. Moreover, GDP also generalizes well for natural images or synthesized images with arbitrary sizes from various tasks out of the distribution of the ImageNet training set.

cs.CV↗

arXiv2023

Value-added catalog of M-giant stars in LAMOST DR9

In this work, we update the catalog of M-giant stars from the low-resolution spectra of the Large Sky Area Multi-Object Fiber Spectroscopic Telescope (LAMOST) Data Release 9. There are 58,076 M giants identified from the classification pipeline with seven temperature subtypes from M0 to M6. The 2471 misclassified non-M-giant stars are white dwarf binaries, early types, and M dwarfs. And the contamination rate is 4.2$\%$ in the M-giants sample. A total of 372 carbon stars were identified by CaH spectral indices, and were further confirmed by the LAMOST spectra. We update the correlation between the $(W1-W2)_0$ color and [M/H] from APOGEE DR17. We calculate the radial velocities of all M giants by applying cross-correlation to the spectra between 8000 and 8950 Åwith synthetic spectra from ATLAS9. Taking star distances less than 4 kpc from Gaia EDR3 as the standard, we refitted the photometric distance relation of M giants. And based on our M-giant stars, we select a group of Sagittarius stream members, whose sky and 3D velocity distributions are well consistent with K-giant Saggitarius stream members found in Yang et al. With our M giants, we find that the disk is asymmetric out to R = 25 kpc, which is 5 kpc further out than detected using K giants.

astro-ph.SR↗

arXiv2023

A Closer Look at Few-Shot 3D Point Cloud Classification

In recent years, research on few-shot learning (FSL) has been fast-growing in the 2D image domain due to the less requirement for labeled training data and greater generalization for novel classes. However, its application in 3D point cloud data is relatively under-explored. Not only need to distinguish unseen classes as in the 2D domain, 3D FSL is more challenging in terms of irregular structures, subtle inter-class differences, and high intra-class variances {when trained on a low number of data.} Moreover, different architectures and learning algorithms make it difficult to study the effectiveness of existing 2D FSL algorithms when migrating to the 3D domain. In this work, for the first time, we perform systematic and extensive investigations of directly applying recent 2D FSL works to 3D point cloud related backbone networks and thus suggest a strong learning baseline for few-shot 3D point cloud classification. Furthermore, we propose a new network, Point-cloud Correlation Interaction (PCIA), with three novel plug-and-play components called Salient-Part Fusion (SPF) module, Self-Channel Interaction Plus (SCI+) module, and Cross-Instance Fusion Plus (CIF+) module to obtain more representative embeddings and improve the feature distinction. These modules can be inserted into most FSL algorithms with minor changes and significantly improve the performance. Experimental results on three benchmark datasets, ModelNet40-FS, ShapeNet70-FS, and ScanObjectNN-FS, demonstrate that our method achieves state-of-the-art performance for the 3D FSL task. Code and datasets are available at https://github.com/cgye96/A_Closer_Look_At_3DFSL.

cs.CV↗

arXiv2023

Switching Pushing Skill Combined MPC and Deep Reinforcement Learning for Planar Non-prehensile Manipulation

In this paper, a novel switching pushing skill algorithm is proposed to improve the efficiency of planar non-prehensile manipulation, which draws inspiration from human pushing actions and comprises two sub-problems, i.e., discrete decision-making of pushing point and continuous feedback control of pushing action. In order to solve the sub-problems above, a combination of Model Predictive Control (MPC) and Deep Reinforcement Learning (DRL) method is employed. Firstly, the selection of pushing point is modeled as a Markov decision process,and an off-policy DRL method is used by reshaping the reward function to train the decision-making model for selecting pushing point from a pre-constructed set based on the current state. Secondly, a motion constraint region (MCR) is constructed for the specific pushing point based on the distance from the target, followed by utilizing the MPC controller to regulate the motion of the object within the MCR towards the target pose. The trigger condition for switching the pushing point occurs when the object reaches the boundary of the MCR under the pushing action. Subsequently, the pushing point and the controller are updated iteratively until the target pose is reached. We conducted pushing experiments on four distinct object shapes in both simulated and physical environments to evaluate our method. The results indicate that our method achieves a significantly higher training efficiency, with a training time that is only about 20% of the baseline method while maintaining around the same success rate. Moreover, our method outperforms the baseline method in terms of both training and execution efficiency of pushing operations, allowing for rapid learning of robot pushing skills.

cs.RO↗

arXiv2023

MetaPortrait: Identity-Preserving Talking Head Generation with Fast Personalized Adaptation

In this work, we propose an ID-preserving talking head generation framework, which advances previous methods in two aspects. First, as opposed to interpolating from sparse flow, we claim that dense landmarks are crucial to achieving accurate geometry-aware flow fields. Second, inspired by face-swapping methods, we adaptively fuse the source identity during synthesis, so that the network better preserves the key characteristics of the image portrait. Although the proposed model surpasses prior generation fidelity on established benchmarks, to further make the talking head generation qualified for real usage, personalized fine-tuning is usually needed. However, this process is rather computationally demanding that is unaffordable to standard users. To solve this, we propose a fast adaptation model using a meta-learning approach. The learned model can be adapted to a high-quality personalized model as fast as 30 seconds. Last but not the least, a spatial-temporal enhancement module is proposed to improve the fine details while ensuring temporal coherency. Extensive experiments prove the significant superiority of our approach over the state of the arts in both one-shot and personalized settings.

cs.CV↗

arXiv2023

Pi-ViMo: Physiology-inspired Robust Vital Sign Monitoring using mmWave Radars

Continuous monitoring of human vital signs using non-contact mmWave radars is attractive due to their ability to penetrate garments and operate under different lighting conditions. Unfortunately, most prior research requires subjects to stay at a fixed distance from radar sensors and to remain still during monitoring. These restrictions limit the applications of radar vital sign monitoring in real life scenarios. In this paper, we address these limitations and present "Pi-ViMo", a non-contact Physiology-inspired Robust Vital Sign Monitoring system, using mmWave radars. We first derive a multi-scattering point model for the human body, and introduce a coherent combining of multiple scatterings to enhance the quality of estimated chest-wall movements. It enables vital sign estimations of subjects at any location in a radar's field of view. We then propose a template matching method to extract human vital signs by adopting physical models of respiration and cardiac activities. The proposed method is capable to separate respiration and heartbeat in the presence of micro-level random body movements (RBM) when a subject is at any location within the field of view of a radar. Experiments in a radar testbed show average respiration rate errors of 6% and heart rate errors of 11.9% for the stationary subjects and average errors of 13.5% for respiration rate and 13.6% for heart rate for subjects under different RBMs.

eess.SP↗

arXiv2023

MMFormer: Multimodal Transformer Using Multiscale Self-Attention for Remote Sensing Image Classification

To benefit the complementary information between heterogeneous data, we introduce a new Multimodal Transformer (MMFormer) for Remote Sensing (RS) image classification using Hyperspectral Image (HSI) accompanied by another source of data such as Light Detection and Ranging (LiDAR). Compared with traditional Vision Transformer (ViT) lacking inductive biases of convolutions, we first introduce convolutional layers to our MMFormer to tokenize patches from multimodal data of HSI and LiDAR. Then we propose a Multi-scale Multi-head Self-Attention (MSMHSA) module to address the problem of compatibility which often limits to fuse HSI with high spectral resolution and LiDAR with relatively low spatial resolution. The proposed MSMHSA module can incorporate HSI to LiDAR data in a coarse-to-fine manner enabling us to learn a fine-grained representation. Extensive experiments on widely used benchmarks (e.g., Trento and MUUFL) demonstrate the effectiveness and superiority of our proposed MMFormer for RS image classification.

cs.CV↗

arXiv2023

UMC: A Unified Bandwidth-efficient and Multi-resolution based Collaborative Perception Framework

Multi-agent collaborative perception (MCP) has recently attracted much attention. It includes three key processes: communication for sharing, collaboration for integration, and reconstruction for different downstream tasks. Existing methods pursue designing the collaboration process alone, ignoring their intrinsic interactions and resulting in suboptimal performance. In contrast, we aim to propose a Unified Collaborative perception framework named UMC, optimizing the communication, collaboration, and reconstruction processes with the Multi-resolution technique. The communication introduces a novel trainable multi-resolution and selective-region (MRSR) mechanism, achieving higher quality and lower bandwidth. Then, a graph-based collaboration is proposed, conducting on each resolution to adapt the MRSR. Finally, the reconstruction integrates the multi-resolution collaborative features for downstream tasks. Since the general metric can not reflect the performance enhancement brought by MCP systematically, we introduce a brand-new evaluation metric that evaluates the MCP from different perspectives. To verify our algorithm, we conducted experiments on the V2X-Sim and OPV2V datasets. Our quantitative and qualitative experiments prove that the proposed UMC greatly outperforms the state-of-the-art collaborative perception approaches.

cs.CV↗

arXiv2023

An mCherry biolaser based on microbubble cavity with ultra-low threshold

Biolasers show considerable potential in the biomedical field. Fluorescent protein (FP) is a type of biomaterial with good luminescence efficiency that can be used as the luminescent gain medium in biolasers. Due to the higher cell/tissue permeability, lower cell phototoxicity, and relatively less background fluorescence than other fluorescent proteins, the red fluorescent protein is more suitable in biological applications. MCherry is the most extensively used high-quality red fluorescent protein because of its short maturation time and stable luminescence properties. In this study, using mCherry and microbubble cavity, we realize a highly stable mCherry fluorescent protein laser. The laser resonator achieves a quality factor of 10^8, which is the highest Q factor among the currently available FP lasers. Moreover, this laser exhibits low threshold of 286 fJ, which can effectively protect the luminescent material from being damaged by pump light. Such a threshold is the lowest in the FP lasers as per our knowledge. The prepared laser shows excellent stability in a wide pH range with good photobleaching resistance and can be stored at 4 degree for nearly a month. Also, the laser can serve as a high-sensitivity molecular concentration detector with mCherry as biomarker, owing to its lasing threshold behavior.

physics.optics↗

arXiv2023

Performance-aware Approximation of Global Channel Pruning for Multitask CNNs

Global channel pruning (GCP) aims to remove a subset of channels (filters) across different layers from a deep model without hurting the performance. Previous works focus on either single task model pruning or simply adapting it to multitask scenario, and still face the following problems when handling multitask pruning: 1) Due to the task mismatch, a well-pruned backbone for classification task focuses on preserving filters that can extract category-sensitive information, causing filters that may be useful for other tasks to be pruned during the backbone pruning stage; 2) For multitask predictions, different filters within or between layers are more closely related and interacted than that for single task prediction, making multitask pruning more difficult. Therefore, aiming at multitask model compression, we propose a Performance-Aware Global Channel Pruning (PAGCP) framework. We first theoretically present the objective for achieving superior GCP, by considering the joint saliency of filters from intra- and inter-layers. Then a sequentially greedy pruning strategy is proposed to optimize the objective, where a performance-aware oracle criterion is developed to evaluate sensitivity of filters to each task and preserve the globally most task-related filters. Experiments on several multitask datasets show that the proposed PAGCP can reduce the FLOPs and parameters by over 60% with minor performance drop, and achieves 1.2x$\sim$3.3x acceleration on both cloud and mobile platforms.

cs.CV↗

arXiv2023

The parallax and 3D kinematics of water masers in the massive star-forming region G034.43+0.24

We report a trigonometric parallax measurement of 22 GHz water masers in the massive star-forming region G034.43+0.24 as part of the Bar and Spiral Structure Legacy (BeSSeL) Survey using the Very Long Baseline Array. The parallax is 0.330$\pm$50.018 mas, corresponding to a distance of $3.03^{+0.17}_{-0.16}$ kpc. This locates G034.43+0.24 near the inner edge of the Sagittarius spiral arm and at one end of a linear distribution of massive young stars which cross nearly the full width of the arm. The measured 3-dimensional motion of G034.43+0.24 indicates a near-circular Galactic orbit. The water masers display arc-like distributions, possibly bow shocks, associated with winds from one or more massive young stars.

astro-ph.GA↗

arXiv2023

Atmospheric parameters and kinematic information for the M giants stars from LAMOST DR9

A catalog of more than 43,000 M giant stars has been selected by Li et al. from the ninth data release of LAMOST. Using the data-driven method SLAM, we obtain the stellar parameters (Teff, logg, [M/H], [$α$/M]) for all the M giant stars with uncertainties of 57 K, 0.25 dex, 0.16 dex and 0.06 dex at SNR > 100, respectively. With those stellar parameters, we constrain the absolute magnitude in K-band, which brings distance with relative uncertainties around 25% statistically. Radial velocities are also calculated by applying cross correlation on the spectra between 8000 A $Å$ and 8950 A $Å$ with synthetic spectra from ATLAS9, which covers the Ca II triplet. Comparison between our radial velocities and those from APOGEE DR17 and Gaia DR3 shows that our radial velocities have a system offset and dispersion around 1 and 4.6 km s$^{-1}$, respectively. With the distances and radial velocities combining with the astrometric data from Gaia DR3, we calculate the full 6D position and velocity information, which are able to be used for further chemo-dynamic studies on the disk and substructures in the halo, especially the Sagittarius Stream.

astro-ph.SR↗

arXiv2023

UniDA3D: Unified Domain Adaptive 3D Semantic Segmentation Pipeline

State-of-the-art 3D semantic segmentation models are trained on off-the-shelf public benchmarks, but they will inevitably face the challenge of recognition accuracy drop when these well-trained models are deployed to a new domain. In this paper, we introduce a Unified Domain Adaptive 3D semantic segmentation pipeline (UniDA3D) to enhance the weak generalization ability, and bridge the point distribution gap between domains. Different from previous studies that only focus on a single adaptation task, UniDA3D can tackle several adaptation tasks in 3D segmentation field, by designing a unified source-and-target active sampling strategy, which selects a maximally-informative subset from both source and target domains for effective model adaptation. Besides, benefiting from the rise of multi-modal 2D-3D datasets, UniDA3D investigates the possibility of achieving a multi-modal sampling strategy, by developing a cross-modality feature interaction module that can extract a representative pair of image and point features to achieve a bi-directional image-point feature interaction for safe model adaptation. Experimentally, UniDA3D is verified to be effective in many adaptation tasks including: 1) unsupervised domain adaptation, 2) unsupervised few-shot domain adaptation; 3) active domain adaptation. Their results demonstrate that, by easily coupling UniDA3D with off-the-shelf 3D segmentation baselines, domain generalization ability of these baselines can be enhanced.

cs.CV↗

arXiv2023

Bi3D: Bi-domain Active Learning for Cross-domain 3D Object Detection

Unsupervised Domain Adaptation (UDA) technique has been explored in 3D cross-domain tasks recently. Though preliminary progress has been made, the performance gap between the UDA-based 3D model and the supervised one trained with fully annotated target domain is still large. This motivates us to consider selecting partial-yet-important target data and labeling them at a minimum cost, to achieve a good trade-off between high performance and low annotation cost. To this end, we propose a Bi-domain active learning approach, namely Bi3D, to solve the cross-domain 3D object detection task. The Bi3D first develops a domainness-aware source sampling strategy, which identifies target-domain-like samples from the source domain to avoid the model being interfered by irrelevant source data. Then a diversity-based target sampling strategy is developed, which selects the most informative subset of target domain to improve the model adaptability to the target domain using as little annotation budget as possible. Experiments are conducted on typical cross-domain adaptation scenarios including cross-LiDAR-beam, cross-country, and cross-sensor, where Bi3D achieves a promising target-domain detection accuracy (89.63% on KITTI) compared with UDAbased work (84.29%), even surpassing the detector trained on the full set of the labeled target domain (88.98%). Our code is available at: https://github.com/PJLabADG/3DTrans.

cs.CV↗

arXiv2023

Understanding Adversarial Attacks on Observations in Deep Reinforcement Learning

Deep reinforcement learning models are vulnerable to adversarial attacks that can decrease a victim's cumulative expected reward by manipulating the victim's observations. Despite the efficiency of previous optimization-based methods for generating adversarial noise in supervised learning, such methods might not be able to achieve the lowest cumulative reward since they do not explore the environmental dynamics in general. In this paper, we provide a framework to better understand the existing methods by reformulating the problem of adversarial attacks on reinforcement learning in the function space. Our reformulation generates an optimal adversary in the function space of the targeted attacks, repelling them via a generic two-stage framework. In the first stage, we train a deceptive policy by hacking the environment, and discover a set of trajectories routing to the lowest reward or the worst-case performance. Next, the adversary misleads the victim to imitate the deceptive policy by perturbing the observations. Compared to existing approaches, we theoretically show that our adversary is stronger under an appropriate noise level. Extensive experiments demonstrate our method's superiority in terms of efficiency and effectiveness, achieving the state-of-the-art performance in both Atari and MuJoCo environments.

cs.LG↗

arXiv2023

The NTSC VLBI System and its application in UT1 measurement

In order to measure the Universal Time (UT1) in real time, National Time Service Center (NTSC) has built a VGOS-like (VLBI Global Observing System) broadband VLBI network, which includes three 13-m radio telescopes located in Jilin, Sanya and Kashi, and a data analysis center in Xi'an. Each station is equipped with a highly stable hydrogen atomic clock and a self-developed VLBI backend, and is co-located with two GPS receivers. This VGOS-like VLBI network may play an important role in improving the Chinese broadband VLBI technology and making valuable contributions to domestic VLBI measurements of UT1. In this paper, we introduce the specifications of this VLBI network, and present the UT1 measurements at C-band conducted in 2018 using the Jilin-Kashi baseline of this network. The comparisons between our UT1 estimates and those provided by IERS suggest that the NTSC VLBI network is capable to determine UT1 accurate at the level of 58.8 microseconds.

astro-ph.IM↗