arXiv subjects

Jianwei Lu

Publications and source records attributed to Jianwei Lu.

12 recordsLinked to original sources

arXiv2025

Every Subtlety Counts: Fine-grained Person Independence Micro-Action Recognition via Distributionally Robust Optimization

Micro-action Recognition is vital for psychological assessment and human-computer interaction. However, existing methods often fail in real-world scenarios because inter-person variability causes the same action to manifest differently, hindering robust generalization. To address this, we propose the Person Independence Universal Micro-action Recognition Framework, which integrates Distributionally Robust Optimization principles to learn person-agnostic representations. Our framework contains two plug-and-play components operating at the feature and loss levels. At the feature level, the Temporal-Frequency Alignment Module normalizes person-specific motion characteristics with a dual-branch design: the temporal branch applies Wasserstein-regularized alignment to stabilize dynamic trajectories, while the frequency branch introduces variance-guided perturbations to enhance robustness against person-specific spectral differences. A consistency-driven fusion mechanism integrates both branches. At the loss level, the Group-Invariant Regularized Loss partitions samples into pseudo-groups to simulate unseen person-specific distributions. By up-weighting boundary cases and regularizing subgroup variance, it forces the model to generalize beyond easy or frequent samples, thus enhancing robustness to difficult variations. Experiments on the large-scale MA-52 dataset demonstrate that our framework outperforms existing methods in both accuracy and robustness, achieving stable generalization under fine-grained conditions.

cs.CV↗

arXiv2023

A Random-patch based Defense Strategy Against Physical Attacks for Face Recognition Systems

The physical attack has been regarded as a kind of threat against real-world computer vision systems. Still, many existing defense methods are only useful for small perturbations attacks and can't detect physical attacks effectively. In this paper, we propose a random-patch based defense strategy to robustly detect physical attacks for Face Recognition System (FRS). Different from mainstream defense methods which focus on building complex deep neural networks (DNN) to achieve high recognition rate on attacks, we introduce a patch based defense strategy to a standard DNN aiming to obtain robust detection models. Extensive experimental results on the employed datasets show the superiority of the proposed defense method on detecting white-box attacks and adaptive attacks which attack both FRS and the defense method. Additionally, due to the simpleness yet robustness of our method, it can be easily applied to the real world face recognition system and extended to other defense methods to boost the detection performance.

cs.CV↗

arXiv2022

Improving Long-tailed Object Detection with Image-Level Supervision by Multi-Task Collaborative Learning

Data in real-world object detection often exhibits the long-tailed distribution. Existing solutions tackle this problem by mitigating the competition between the head and tail categories. However, due to the scarcity of training samples, tail categories are still unable to learn discriminative representations. Bringing more data into the training may alleviate the problem, but collecting instance-level annotations is an excruciating task. In contrast, image-level annotations are easily accessible but not fully exploited. In this paper, we propose a novel framework CLIS (multi-task Collaborative Learning with Image-level Supervision), which leverage image-level supervision to enhance the detection ability in a multi-task collaborative way. Specifically, there are an object detection task (consisting of an instance-classification task and a localization task) and an image-classification task in our framework, responsible for utilizing the two types of supervision. Different tasks are trained collaboratively by three key designs: (1) task-specialized sub-networks that learn specific representations of different tasks without feature entanglement. (2) a siamese sub-network for the image-classification task that shares its knowledge with the instance-classification task, resulting in feature enrichment of detectors. (3) a contrastive learning regularization that maintains representation consistency, bridging feature gaps of different supervision. Extensive experiments are conducted on the challenging LVIS dataset. Without sophisticated loss engineering, CLIS achieves an overall AP of 31.1 with 10.1 point improvement on tail categories, establishing a new state-of-the-art. Code will be at https://github.com/waveboo/CLIS.

cs.CV↗

arXiv2022

Equalized Focal Loss for Dense Long-Tailed Object Detection

Despite the recent success of long-tailed object detection, almost all long-tailed object detectors are developed based on the two-stage paradigm. In practice, one-stage detectors are more prevalent in the industry because they have a simple and fast pipeline that is easy to deploy. However, in the long-tailed scenario, this line of work has not been explored so far. In this paper, we investigate whether one-stage detectors can perform well in this case. We discover the primary obstacle that prevents one-stage detectors from achieving excellent performance is: categories suffer from different degrees of positive-negative imbalance problems under the long-tailed data distribution. The conventional focal loss balances the training process with the same modulating factor for all categories, thus failing to handle the long-tailed problem. To address this issue, we propose the Equalized Focal Loss (EFL) that rebalances the loss contribution of positive and negative samples of different categories independently according to their imbalance degrees. Specifically, EFL adopts a category-relevant modulating factor which can be adjusted dynamically by the training status of different categories. Extensive experiments conducted on the challenging LVIS v1 benchmark demonstrate the effectiveness of our proposed method. With an end-to-end training pipeline, EFL achieves 29.2% in terms of overall AP and obtains significant performance improvements on rare categories, surpassing all existing state-of-the-art methods. The code is available at https://github.com/ModelTC/EOD.

cs.CV↗

arXiv2021

PoissonSeg: Semi-Supervised Few-Shot Medical Image Segmentation via Poisson Learning

The application of deep learning to medical image segmentation has been hampered due to the lack of abundant pixel-level annotated data. Few-shot Semantic Segmentation (FSS) is a promising strategy for breaking the deadlock. However, a high-performing FSS model still requires sufficient pixel-level annotated classes for training to avoid overfitting, which leads to its performance bottleneck in medical image segmentation due to the unmet need for annotations. Thus, semi-supervised FSS for medical images is accordingly proposed to utilize unlabeled data for further performance improvement. Nevertheless, existing semi-supervised FSS methods has two obvious defects: (1) neglecting the relationship between the labeled and unlabeled data; (2) using unlabeled data directly for end-to-end training leads to degenerated representation learning. To address these problems, we propose a novel semi-supervised FSS framework for medical image segmentation. The proposed framework employs Poisson learning for modeling data relationship and propagating supervision signals, and Spatial Consistency Calibration for encouraging the model to learn more coherent representations. In this process, unlabeled samples do not involve in end-to-end training, but provide supervisory information for query image segmentation through graph-based learning. We conduct extensive experiments on three medical image segmentation datasets (i.e. ISIC skin lesion segmentation, abdominal organs segmentation for MRI and abdominal organs segmentation for CT) to demonstrate the state-of-the-art performance and broad applicability of the proposed framework.

cs.CV↗

arXiv2021

L-SNet: from Region Localization to Scale Invariant Medical Image Segmentation

Coarse-to-fine models and cascade segmentation architectures are widely adopted to solve the problem of large scale variations in medical image segmentation. However, those methods have two primary limitations: the first-stage segmentation becomes a performance bottleneck; the lack of overall differentiability makes the training process of two stages asynchronous and inconsistent. In this paper, we propose a differentiable two-stage network architecture to tackle these problems. In the first stage, a localization network (L-Net) locates Regions of Interest (RoIs) in a detection fashion; in the second stage, a segmentation network (S-Net) performs fine segmentation on the recalibrated RoIs; a RoI recalibration module between L-Net and S-Net eliminating the inconsistencies. Experimental results on the public dataset show that our method outperforms state-of-the-art coarse-to-fine models with negligible computation overheads.

cs.CV↗

arXiv2020

Cross-Modal Self-Attention Distillation for Prostate Cancer Segmentation

Automatic segmentation of the prostate cancer from the multi-modal magnetic resonance images is of critical importance for the initial staging and prognosis of patients. However, how to use the multi-modal image features more efficiently is still a challenging problem in the field of medical image segmentation. In this paper, we develop a cross-modal self-attention distillation network by fully exploiting the encoded information of the intermediate layers from different modalities, and the extracted attention maps of different modalities enable the model to transfer the significant spatial information with more details. Moreover, a novel spatial correlated feature fusion module is further employed for learning more complementary correlation and non-linear information of different modality images. We evaluate our model in five-fold cross-validation on 358 MRI with biopsy confirmed. Extensive experiment results demonstrate that our proposed network achieves state-of-the-art performance.

eess.IV↗

arXiv2017

Probing the Chiral Anomaly by Planar Hall Effect in Three-dimensional Dirac Semimetal Cd3As2 Nanoplates

Searching for exotic transport properties in new topological state of matters is an active topic. One of the most fascinating achievements is the chiral anomaly in recently discovered Weyl semimetals (WSMs), which is manifested as a negative longitudinal magnetoresistance (LMR) in the presence of a magnetic field B parallel to an electric field E. Another predicted key effect closely related to the chiral anomaly is the planar Hall effect (PHE), which has not been identified in WSMs so far. Here we carried out the planar Hall measurements on Cd3As2 nanoplates, and found that, accompanied by the large negative LMR, a PHE with non-zero transverse voltage can be developed while tilting the in-plane magnetic field B away from the electric field E. Further experiments reveal that both the PHE and the negative LMR can be suppressed synchronously by increasing the temperature, but still visible at room temperature, indicating the same origin of these two effects. The observation of PHE in Cd3As2 nanoplates gives another transport evidence for the chiral anomaly and provides a deep insight into the chiral charge pumping in Weyl Fermions system.

cond-mat.mes-hall↗

arXiv2017

Field induced topological phase transition from a three-dimensional Weyl semimetal to a two-dimensional massive Dirac metal in ZrTe5

Symmetry protected Dirac semimetals can be transformed into Weyl semimetals by breaking the protecting symmetry, leading to many exotic quantum phenomena such as chiral anomaly and anomalous Hall effect. Here we show that, due to the large Zeeman g factor and small band width along b-axis in Dirac semimetal ZrTe5, a magnetic field of about 8 T along b-axis direction may annihilate the Weyl points and open up a two-dimensional (2D) Dirac mass gap, when the Zeeman splitting exceeds the band width along b-axis. This is manifested by a sharp drop of magnetoresistance (MR) above 8 T, which is probably due to additional carriers induced by the orbital splitting of the zeroth Landau level associated with the 2D Dirac point, which is descendant of the original Weyl points. Further evidence of the additional carriers is provided by the Hall effect and different anisotropic magnetoresistance (AMR) in low and high field regions. Our experiment reveals a probable topological quantum phase transition of field induced Weyl points annihilation in Dirac semimetal ZrTe5 and gives an alternative explanation for the drop of MR at high field.

cond-mat.mtrl-sci↗

arXiv2016

Thickness-tuned transition of band topology in ZrTe5 nanosheets

We report thickness-tuned electrical transport and Hall resistivity in highly anisotropic three-dimensional Dirac semimetal ZrTe5 nanosheets. We find that when the thickness of the nanosheet is blow about 40 nm, the system takes a clear transition from topological semimetal with two bands carriers to a single band with conventional hole carriers. The resistivity peak temperature T* decreases systematically with decreasing thickness down to about 40 nm, then shifts up with the further decrease of the thickness. Analysis of the data below 40 nm indicates that the hole carriers completely dominate the transport in the entire temperature range, regardless of the temperature being below or above T*. By further tracking the carrier density, we find that the Fermi level shifts consecutively downward from conduction band to the valence band as decreasing the thickness. Our experiments unambiguously reveal a highly thickness-tuned transition of band topology in ZrTe5 nanosheets.

cond-mat.mes-hall↗

arXiv2016

Detecting "Smart" Spammers On Social Network: A Topic Model Approach

Spammer detection on social network is a challenging problem. The rigid anti-spam rules have resulted in emergence of "smart" spammers. They resemble legitimate users who are difficult to identify. In this paper, we present a novel spammer classification approach based on Latent Dirichlet Allocation(LDA), a topic model. Our approach extracts both the local and the global information of topic distribution patterns, which capture the essence of spamming. Tested on one benchmark dataset and one self-collected dataset, our proposed method outperforms other state-of-the-art methods in terms of averaged F1-score.

cs.CL↗

arXiv2016

Transport evidence for the three-dimensional Dirac semimetal phase in ZrTe5

Topological Dirac semimetal is a newly discovered class of materials and has attracted intense attentions. This material can be viewed as a three-dimensional (3D) analogue of graphene and has linear energy dispersion in bulk, leading to a range of exotic transport properties. Here we report direct quantum transport evidence of 3D Dirac semimetal phase of layered material ZrTe5 by angular dependent magnetoresistance measurements under high magnetic fields up to 31 Tesla. We observed very clear negative longitudinal magnetoresistance induced by chiral anomaly under the condition of the magnetic field aligned only along the current direction. Pronounced Shubnikov-de Hass (SdH) quantum oscillations in both longitudinal magnetoresistance and transverse Hall resistance were observed, revealing anisotropic light cyclotron masses and high mobility of the system. In particular, a nontrivial π-Berry phase in the SdH gives clear evidence for 3D Dirac semimetal phase. Furthermore, we observed clear Landau Level splitting under high magnetic field, suggesting possible splitting of Dirac point into Weyl points due to broken time reversal symmetry. Our results indicate that ZrTe5 is an ideal platform to study 3D massless Dirac and Weyl fermions in a layered compound.

cond-mat.mes-hall↗