arXiv subjects

Na Zhao

Publications and source records attributed to Na Zhao.

At least 73 records · Page 4Linked to original sources

arXiv2024

CT3D++: Improving 3D Object Detection with Keypoint-induced Channel-wise Transformer

The field of 3D object detection from point clouds is rapidly advancing in computer vision, aiming to accurately and efficiently detect and localize objects in three-dimensional space. Current 3D detectors commonly fall short in terms of flexibility and scalability, with ample room for advancements in performance. In this paper, our objective is to address these limitations by introducing two frameworks for 3D object detection with minimal hand-crafted design. Firstly, we propose CT3D, which sequentially performs raw-point-based embedding, a standard Transformer encoder, and a channel-wise decoder for point features within each proposal. Secondly, we present an enhanced network called CT3D++, which incorporates geometric and semantic fusion-based embedding to extract more valuable and comprehensive proposal-aware information. Additionally, CT3D ++ utilizes a point-to-key bidirectional encoder for more efficient feature encoding with reduced computational cost. By replacing the corresponding components of CT3D with these novel modules, CT3D++ achieves state-of-the-art performance on both the KITTI dataset and the large-scale Way\-mo Open Dataset. The source code for our frameworks will be made accessible at https://github.com/hlsheng1/CT3D-plusplus.

cs.CV↗

arXiv2024

Improving 3D Occupancy Prediction through Class-balancing Loss and Multi-scale Representation

3D environment recognition is essential for autonomous driving systems, as autonomous vehicles require a comprehensive understanding of surrounding scenes. Recently, the predominant approach to define this real-life problem is through 3D occupancy prediction. It attempts to predict the occupancy states and semantic labels for all voxels in 3D space, which enhances the perception capability. Birds-Eye-View(BEV)-based perception has achieved the SOTA performance for this task. Nonetheless, this architecture fails to represent various scales of BEV features. In this paper, inspired by the success of UNet in semantic segmentation tasks, we introduce a novel UNet-like Multi-scale Occupancy Head module to relieve this issue. Furthermore, we propose the class-balancing loss to compensate for rare classes in the dataset. The experimental results on nuScenes 3D occupancy challenge dataset show the superiority of our proposed approach over baseline and SOTA methods.

cs.CV↗

arXiv2024

Performance-lossless Black-box Model Watermarking

With the development of deep learning, high-value and high-cost models have become valuable assets, and related intellectual property protection technologies have become a hot topic. However, existing model watermarking work in black-box scenarios mainly originates from training-based backdoor methods, which probably degrade primary task performance. To address this, we propose a branch backdoor-based model watermarking protocol to protect model intellectual property, where a construction based on a message authentication scheme is adopted as the branch indicator after a comparative analysis with secure cryptographic technologies primitives. We prove the lossless performance of the protocol by reduction. In addition, we analyze the potential threats to the protocol and provide a secure and feasible watermarking instance for language models.

cs.CR↗

arXiv2024

Dual-Perspective Knowledge Enrichment for Semi-Supervised 3D Object Detection

Semi-supervised 3D object detection is a promising yet under-explored direction to reduce data annotation costs, especially for cluttered indoor scenes. A few prior works, such as SESS and 3DIoUMatch, attempt to solve this task by utilizing a teacher model to generate pseudo-labels for unlabeled samples. However, the availability of unlabeled samples in the 3D domain is relatively limited compared to its 2D counterpart due to the greater effort required to collect 3D data. Moreover, the loose consistency regularization in SESS and restricted pseudo-label selection strategy in 3DIoUMatch lead to either low-quality supervision or a limited amount of pseudo labels. To address these issues, we present a novel Dual-Perspective Knowledge Enrichment approach named DPKE for semi-supervised 3D object detection. Our DPKE enriches the knowledge of limited training data, particularly unlabeled data, from two perspectives: data-perspective and feature-perspective. Specifically, from the data-perspective, we propose a class-probabilistic data augmentation method that augments the input data with additional instances based on the varying distribution of class probabilities. Our DPKE achieves feature-perspective knowledge enrichment by designing a geometry-aware feature matching method that regularizes feature-level similarity between object proposals from the student and teacher models. Extensive experiments on the two benchmark datasets demonstrate that our DPKE achieves superior performance over existing state-of-the-art approaches under various label ratio conditions. The source code will be made available to the public.

cs.CV↗

arXiv2023

Hydrogel modified evaporation interface for highly stable membrane distillation

Surface effect of low-surface-tension contaminants accumulating at the evaporation surface can easily induce membrane wetting in the application of membrane distillation, especially in hypersaline scenarios. In this work, we propose a novel strategy to eliminate the surface effect and redistribute contaminants at the evaporation interface with simply incorporating a layer of hydrogel. The as-fabricated composite membrane exhibits remarkable stability, even when exposed to extreme conditions, such as a salt concentration of 5M and surfactant concentration of 8 mM. The breakthrough pressure of the membrane is as high as 20 bars in the presence of surfactants, surpassing commercial hydrophobic membranes by one to two magnitudes. Combined study of density functional theory and molecular dynamics simulations reveals the important role of hydrogel-surfactant interaction in suppressing the surface effect. As a proof of concept, we also demonstrate the stable performance of the membrane in processing synthetic wastewater containing surfactants of 144 mg L-1, mineral oils of 1g L-1 and NaCl of 192 g L-1, showing potential of the membrane in addressing challenges of hypersaline water treatment and zero liquid discharge processes.

physics.chem-ph↗

arXiv2023

Style-Hallucinated Dual Consistency Learning: A Unified Framework for Visual Domain Generalization

Domain shift widely exists in the visual world, while modern deep neural networks commonly suffer from severe performance degradation under domain shift due to the poor generalization ability, which limits the real-world applications. The domain shift mainly lies in the limited source environmental variations and the large distribution gap between source and unseen target data. To this end, we propose a unified framework, Style-HAllucinated Dual consistEncy learning (SHADE), to handle such domain shift in various visual tasks. Specifically, SHADE is constructed based on two consistency constraints, Style Consistency (SC) and Retrospection Consistency (RC). SC enriches the source situations and encourages the model to learn consistent representation across style-diversified samples. RC leverages general visual knowledge to prevent the model from overfitting to source data and thus largely keeps the representation consistent between the source and general visual models. Furthermore, we present a novel style hallucination module (SHM) to generate style-diversified samples that are essential to consistency learning. SHM selects basis styles from the source distribution, enabling the model to dynamically generate diverse and realistic samples during training. Extensive experiments demonstrate that our versatile SHADE can significantly enhance the generalization in various visual recognition tasks, including image classification, semantic segmentation and object detection, with different models, i.e., ConvNets and Transformer.

cs.CV↗

arXiv2023

On the energy method for the global solutions to the three dimensional incompressible non-resistive MHD near equilibrium

We prove the global existence of the smooth solutions near equilibrium to the Cauchy problem of the incompressible non-resistive magnetohydrodynamic equations in the whole three dimensional space under some admissible condition. The result has been obtained by Xu and Zhang (SIAM J. Math. Anal. 47: 26--65, 2015) in anisotropic Besov space framework. In this paper, we provide a new proof based on the temporal weighted energy method.

math.AP↗

arXiv2023

Generalized Few-Shot Point Cloud Segmentation Via Geometric Words

Existing fully-supervised point cloud segmentation methods suffer in the dynamic testing environment with emerging new classes. Few-shot point cloud segmentation algorithms address this problem by learning to adapt to new classes at the sacrifice of segmentation accuracy for the base classes, which severely impedes its practicality. This largely motivates us to present the first attempt at a more practical paradigm of generalized few-shot point cloud segmentation, which requires the model to generalize to new categories with only a few support point clouds and simultaneously retain the capability to segment base classes. We propose the geometric words to represent geometric components shared between the base and novel classes, and incorporate them into a novel geometric-aware semantic representation to facilitate better generalization to the new classes without forgetting the old ones. Moreover, we introduce geometric prototypes to guide the segmentation with geometric prior knowledge. Extensive experiments on S3DIS and ScanNet consistently illustrate the superior performance of our method over baseline methods. Our code is available at: https://github.com/Pixie8888/GFS-3DSeg_GWs.

cs.CV↗

arXiv2023

Towards Robust Few-shot Point Cloud Semantic Segmentation

Few-shot point cloud semantic segmentation aims to train a model to quickly adapt to new unseen classes with only a handful of support set samples. However, the noise-free assumption in the support set can be easily violated in many practical real-world settings. In this paper, we focus on improving the robustness of few-shot point cloud segmentation under the detrimental influence of noisy support sets during testing time. To this end, we first propose a Component-level Clean Noise Separation (CCNS) representation learning to learn discriminative feature representations that separates the clean samples of the target classes from the noisy samples. Leveraging the well separated clean and noisy support samples from our CCNS, we further propose a Multi-scale Degree-based Noise Suppression (MDNS) scheme to remove the noisy shots from the support set. We conduct extensive experiments on various noise settings on two benchmark datasets. Our results show that the combination of CCNS and MDNS significantly improves the performance. Our code is available at https://github.com/Pixie8888/R3DFSSeg.

cs.CV↗

arXiv2023

Refining 6-DoF Grasps with Context-Specific Classifiers

In this work, we present GraspFlow, a refinement approach for generating context-specific grasps. We formulate the problem of grasp synthesis as a sampling problem: we seek to sample from a context-conditioned probability distribution of successful grasps. However, this target distribution is unknown. As a solution, we devise a discriminator gradient-flow method to evolve grasps obtained from a simpler distribution in a manner that mimics sampling from the desired target distribution. Unlike existing approaches, GraspFlow is modular, allowing grasps that satisfy multiple criteria to be obtained simply by incorporating the relevant discriminators. It is also simple to implement, requiring minimal code given existing auto-differentiation libraries and suitable discriminators. Experiments show that GraspFlow generates stable and executable grasps on a real-world Panda robot for a diverse range of objects. In particular, in 60 trials on 20 different household objects, the first attempted grasp was successful 94% of the time, and 100% grasp success was achieved by the second grasp. Moreover, incorporating a functional discriminator for robot-human handover improved the functional aspect of the grasp by up to 33%.

cs.RO↗

arXiv2023

On the Performance Tradeoff of an ISAC System with Finite Blocklength

Integrated sensing and communication (ISAC) has been proposed as a promising paradigm in the future wireless networks, where the spectral and hardware resources are shared to provide a considerable performance gain. It is essential to understand how sensing and communication (S\&C) influences each other to guide the practical algorithm and system design in ISAC. In this paper, we investigate the performance tradeoff between S\&C in a single-input single-output (SISO) ISAC system with finite blocklength. In particular, we present the system model and the ISAC scheme, after which the rate-error tradeoff is introduced as the performance metric. Then we derive the achievability and converse bounds for the rate-error tradeoff, determining the boundary of the joint S\&C performance. Furthermore, we develop the asymptotic analysis at large blocklength regime, where the performance tradeoff between S\&C is proved to vanish as the blocklength tends to infinity. Finally, our theoretical analysis is consolidated by simulation results.

cs.IT↗

arXiv2023

School Bullying Results in Poor Psychological Conditions: Evidence from a Survey of 95,545 Subjects

To investigate whether bullying and psychological conditions are correlated, this study analyzed a survey of primary and secondary school students from Zigong City, Sichuan Province. A total of 95,545 students completed a personal information questionnaire, the Multidimensional Peer-Victimization Scale (MPVS), and eight other scales pertaining to various psychological problems. The data showed that 68,315 (71.5\%) participants experienced school bullying at varying degrees, indicating the prevalence of bullying among adolescents. The chi-square tests revealed a strong correlation between school bullying and psychological conditions. This correlation was further explored through multivariate logistic regression, showing that students who experienced mild bullying had a 3.10 times higher probability of emotional and behavioral problems, 4.06 times higher probability of experiencing prodromal symptoms of mental illness, 4.72 times higher probability of anxiety, 3.28 times higher probability of developing post-traumatic stress disorder (PTSD) , 4.07 times higher probability of poor sleep quality, 3.13 times higher probability of internet addiction, 2.18 times higher probability of poor mental health, and 3.64 times higher probability of depression than students who did not experience bullying. The corresponding probabilities for students who experienced severe bullying were 11.35, 17.35, 18.52, 12.59, 11.67, 12.03, 4.64, and 5.34 times higher, respectively. In conclusion, school bullying and psychological conditions are significantly correlated among primary and secondary school students, and the more severe the bullying, the higher the probability to suffer from psychological problems.

physics.soc-ph↗

arXiv2023

Finite speed axially symmetric Navier-Stokes flows passing a cone

Let $D$ be the exterior of a cone inside a ball, with its altitude angle at most $π/6$ in $\mathbb{R}^3$, which touches the $x_3$ axis at the origin. For any initial value $v_0 = v_{0,r}e_{r} + v_{0,θ} e_θ + v_{0,3} e_{3}$ in a $C^2(\overline{D})$ class, which has the usual even-odd-odd symmetry in the $x_3$ variable and has the partial smallness only in the swirl direction: $ | r v_{0, θ} | \leq \frac{1}{100}$, the axially symmetric Navier-Stokes equations (ASNS) with Navier-Hodge-Lions slip boundary condition has a finite-energy solution that stays bounded for all time. In particular, no finite-time blowup of the fluid velocity occurs. Compared with standard smallness assumptions on the initial velocity, no size restriction is made on the components $v_{0,r}$ and $v_{0,3}$. In a broad sense, this result appears to solve $2/3$ of the regularity problem of ASNS in such domains in the class of solutions with the above symmetry. Equivalently, this result is connected to the general open question which asks that if an absolute smallness of one component of the initial velocity implies the global smoothness, see e.g. page 873 in \cite{CZZ17}. Our result seems to give a positive answer in a special setting. As a byproduct, we also construct an unbounded solution of the forced Navier Stokes equation in a special cusp domain that has finite energy. The forcing term, with the scaling factor of $-1$, is in the standard regularity class. This result confirms the intuition that if the channel of a fluid is very thin, arbitrarily high speed in the classical sense can be attained under a mildly singular force which is physically reasonable in view that Newtonian gravity and Coulomb force have scaling factor $-2$.

math.AP↗

arXiv2022

Synthetic-to-Real Domain Generalized Semantic Segmentation for 3D Indoor Point Clouds

Semantic segmentation in 3D indoor scenes has achieved remarkable performance under the supervision of large-scale annotated data. However, previous works rely on the assumption that the training and testing data are of the same distribution, which may suffer from performance degradation when evaluated on the out-of-distribution scenes. To alleviate the annotation cost and the performance degradation, this paper introduces the synthetic-to-real domain generalization setting to this task. Specifically, the domain gap between synthetic and real-world point cloud data mainly lies in the different layouts and point patterns. To address these problems, we first propose a clustering instance mix (CINMix) augmentation technique to diversify the layouts of the source data. In addition, we augment the point patterns of the source data and introduce non-parametric multi-prototypes to ameliorate the intra-class variance enlarged by the augmented point patterns. The multi-prototypes can model the intra-class variance and rectify the global classifier in both training and inference stages. Experiments on the synthetic-to-real benchmark demonstrate that both CINMix and multi-prototypes can narrow the distribution gap and thus improve the generalization ability on real-world datasets.

cs.CV↗

arXiv2022

Myopia prediction for adolescents via time-aware deep learning

Background: Quantitative prediction of the adolescents' spherical equivalent based on their variable-length historical vision records. Methods: From October 2019 to March 2022, we examined binocular uncorrected visual acuity, axial length, corneal curvature, and axial of 75,172 eyes from 37,586 adolescents aged 6-20 years in Chengdu, China. 80\% samples consist of the training set and the remaining 20\% form the testing set. Time-Aware Long Short-Term Memory was used to quantitatively predict the adolescents' spherical equivalent within two and a half years. Result: The mean absolute prediction error on the testing set was 0.273-0.257 for spherical equivalent, ranging from 0.189-0.160 to 0.596-0.473 if we consider different lengths of historical records and different prediction durations. Conclusions: Time-Aware Long Short-Term Memory was applied to captured the temporal features in irregularly sampled time series, which is more in line with the characteristics of real data and thus has higher applicability, and helps to identify the progression of myopia earlier. The overall error 0.273 is much smaller than the criterion for clinically acceptable prediction, say 0.75.

cs.LG↗

arXiv2022

Rethinking IoU-based Optimization for Single-stage 3D Object Detection

Since Intersection-over-Union (IoU) based optimization maintains the consistency of the final IoU prediction metric and losses, it has been widely used in both regression and classification branches of single-stage 2D object detectors. Recently, several 3D object detection methods adopt IoU-based optimization and directly replace the 2D IoU with 3D IoU. However, such a direct computation in 3D is very costly due to the complex implementation and inefficient backward operations. Moreover, 3D IoU-based optimization is sub-optimal as it is sensitive to rotation and thus can cause training instability and detection performance deterioration. In this paper, we propose a novel Rotation-Decoupled IoU (RDIoU) method that can mitigate the rotation-sensitivity issue, and produce more efficient optimization objectives compared with 3D IoU during the training stage. Specifically, our RDIoU simplifies the complex interactions of regression parameters by decoupling the rotation variable as an independent term, yet preserving the geometry of 3D IoU. By incorporating RDIoU into both the regression and classification branches, the network is encouraged to learn more precise bounding boxes and concurrently overcome the misalignment issue between classification and regression. Extensive experiments on the benchmark KITTI and Waymo Open Dataset validate that our RDIoU method can bring substantial improvement for the single-stage 3D object detection.

cs.CV↗

arXiv2022

Style-Hallucinated Dual Consistency Learning for Domain Generalized Semantic Segmentation

In this paper, we study the task of synthetic-to-real domain generalized semantic segmentation, which aims to learn a model that is robust to unseen real-world scenes using only synthetic data. The large domain shift between synthetic and real-world data, including the limited source environmental variations and the large distribution gap between synthetic and real-world data, significantly hinders the model performance on unseen real-world scenes. In this work, we propose the Style-HAllucinated Dual consistEncy learning (SHADE) framework to handle such domain shift. Specifically, SHADE is constructed based on two consistency constraints, Style Consistency (SC) and Retrospection Consistency (RC). SC enriches the source situations and encourages the model to learn consistent representation across style-diversified samples. RC leverages real-world knowledge to prevent the model from overfitting to synthetic data and thus largely keeps the representation consistent between the synthetic and real-world models. Furthermore, we present a novel style hallucination module (SHM) to generate style-diversified samples that are essential to consistency learning. SHM selects basis styles from the source distribution, enabling the model to dynamically generate diverse and realistic samples during training. Experiments show that our SHADE yields significant improvement and outperforms state-of-the-art methods by 5.05% and 8.35% on the average mIoU of three real-world datasets on single- and multi-source settings, respectively.

cs.CV↗

arXiv2022

MRI Reconstruction Using Deep Bayesian Estimation

Purpose: To develop a deep learning-based Bayesian inference for MRI reconstruction. Methods: We modeled the MRI reconstruction problem with Bayes's theorem, following the recently proposed PixelCNN++ method. The image reconstruction from incomplete k-space measurement was obtained by maximizing the posterior possibility. A generative network was utilized as the image prior, which was computationally tractable, and the k-space data fidelity was enforced by using an equality constraint. The stochastic backpropagation was utilized to calculate the descent gradient in the process of maximum a posterior, and a projected subgradient method was used to impose the equality constraint. In contrast to the other deep learning reconstruction methods, the proposed one used the likelihood of prior as the training loss and the objective function in reconstruction to improve the image quality. Results: The proposed method showed an improved performance in preserving image details and reducing aliasing artifacts, compared with GRAPPA, $\ell_1$-ESPRiT, and MODL, a state-of-the-art deep learning reconstruction method. The proposed method generally achieved more than 5 dB peak signal-to-noise ratio improvement for compressed sensing and parallel imaging reconstructions compared with the other methods. Conclusion: The Bayesian inference significantly improved the reconstruction performance, compared with the conventional $\ell_1$-sparsity prior in compressed sensing reconstruction tasks. More importantly, the proposed reconstruction framework can be generalized for most MRI reconstruction scenarios.

cs.CV↗