Q. M. Jonathan Wu

Jiayuan Wang, Q. M. Jonathan Wu, Farhad Pourpanah

Magnetic resonance (MR) and computer tomography (CT) imaging are valuable tools for diagnosing diseases and planning treatment. However, limitations such as radiation exposure and cost can restrict access to certain imaging modalities. To address this issue, medical image synthesis can generate one modality from another, but many existing models struggle with high-quality image synthesis when multiple slices are present in the dataset. This study proposes an attention-based dual contrast generative model, called ADC-cycleGAN, which can synthesize medical images from unpaired data with multiple slices. The model integrates a dual contrast loss term with the CycleGAN loss to ensure that the synthesized images are distinguishable from the source domain. Additionally, an attention mechanism is incorporated into the generators to extract informative features from both channel and spatial domains. To improve performance when dealing with multiple slices, the $K$-means algorithm is used to cluster the dataset into $K$ groups, and each group is used to train a separate ADC-cycleGAN. Experimental results demonstrate that the proposed ADC-cycleGAN model produces comparable samples to other state-of-the-art generative models, achieving the highest PSNR and SSIM values of 19.04385 and 0.68551, respectively. We publish the code at https://github.com/JiayuanWang-JW/ADC-cycleGAN.

Jiayuan Wang, Q. M. Jonathan Wu, Ning Zhang

High precision, lightweight, and real-time responsiveness are three essential requirements for implementing autonomous driving. In this study, we incorporate A-YOLOM, an adaptive, real-time, and lightweight multi-task model designed to concurrently address object detection, drivable area segmentation, and lane line segmentation tasks. Specifically, we develop an end-to-end multi-task model with a unified and streamlined segmentation structure. We introduce a learnable parameter that adaptively concatenates features between necks and backbone in segmentation tasks, using the same loss function for all segmentation tasks. This eliminates the need for customizations and enhances the model's generalization capabilities. We also introduce a segmentation head composed only of a series of convolutional layers, which reduces the number of parameters and inference time. We achieve competitive results on the BDD100k dataset, particularly in visualization outcomes. The performance results show a mAP50 of 81.1% for object detection, a mIoU of 91.0% for drivable area segmentation, and an IoU of 28.8% for lane line segmentation. Additionally, we introduce real-world scenarios to evaluate our model's performance in a real scene, which significantly outperforms competitors. This demonstrates that our model not only exhibits competitive performance but is also more flexible and faster than existing multi-task models. The source codes and pre-trained models are released at https://github.com/JiayuanWang-JW/YOLOv8-multi-task

Jiayuan Wang, Q. M. Jonathan Wu, Ning Zhang et al.

Autonomous driving systems rely on panoptic perception to jointly handle object detection, drivable area segmentation, and lane line segmentation. Although multi-task learning is an effective way to integrate these tasks, its increasing model parameters and complexity make deployment on on-board devices difficult. To address this challenge, we propose a multi-task model compression framework that combines task-aware safe pruning with feature-level knowledge distillation. Our safe pruning strategy integrates Taylor-based channel importance with gradient conflict penalty to keep important channels while removing redundant and conflicting channels. To mitigate performance degradation after pruning, we further design a task head-agnostic distillation method that transfers intermediate backbone and encoder features from a teacher to a student model as guidance. Experiments on the BDD100K dataset demonstrate that our compressed model achieves a 32.7% reduction in parameters while segmentation performance shows negligible accuracy loss and only a minor decrease in detection (-1.2% for Recall and -1.8% for mAP50) compared to the teacher. The compressed model still runs at 32.7 FPS in real-time. These results show that combining pruning and knowledge distillation provides an effective compression solution for multi-task panoptic perception.

Xinyu Wang, Bohan Zhuang, Qi Wu

Large Vision Language Models (LVLMs) have demonstrated remarkable abilities in understanding and reasoning about both visual and textual information. However, existing evaluation methods for LVLMs, primarily based on benchmarks like Visual Question Answering and image captioning, often fail to capture the full scope of LVLMs' capabilities. These benchmarks are limited by issues such as inadequate assessment of detailed visual perception, data contamination, and a lack of focus on multi-turn reasoning. To address these challenges, we propose \method{}, a game-based evaluation framework designed to provide a comprehensive assessment of LVLMs' cognitive and reasoning skills in structured environments. \method{} uses a set of games to evaluate LVLMs on four core tasks: Perceiving, Question Answering, Rule Following, and End-to-End Playing, with each target task designed to assess specific abilities, including visual perception, reasoning, decision-making, etc. Based on this framework, we conduct extensive experiments that explore the limitations of current LVLMs, such as handling long structured outputs and perceiving detailed and dense elements. Code and data are publicly available at https://github.com/xinke-wang/LVLM-Playground.

Yi Li, Zhiyuan Zhang, Jiangnan Xia et al.

This paper presents a novel Two-Stage Diffusion Model (TS-Diff) for enhancing extremely low-light RAW images. In the pre-training stage, TS-Diff synthesizes noisy images by constructing multiple virtual cameras based on a noise space. Camera Feature Integration (CFI) modules are then designed to enable the model to learn generalizable features across diverse virtual cameras. During the aligning stage, CFIs are averaged to create a target-specific CFI$^T$, which is fine-tuned using a small amount of real RAW data to adapt to the noise characteristics of specific cameras. A structural reparameterization technique further simplifies CFI$^T$ for efficient deployment. To address color shifts during the diffusion process, a color corrector is introduced to ensure color consistency by dynamically adjusting global color distributions. Additionally, a novel dataset, QID, is constructed, featuring quantifiable illumination levels and a wide dynamic range, providing a comprehensive benchmark for training and evaluation under extreme low-light conditions. Experimental results demonstrate that TS-Diff achieves state-of-the-art performance on multiple datasets, including QID, SID, and ELD, excelling in denoising, generalization, and color consistency across various cameras and illumination levels. These findings highlight the robustness and versatility of TS-Diff, making it a practical solution for low-light imaging applications. Source codes and models are available at https://github.com/CircccleK/TS-Diff

Dongyang Liu, Peng Gao, David Liu et al.

Diffusion model distillation has emerged as a powerful technique for creating efficient few-step and single-step generators. Among these, Distribution Matching Distillation (DMD) and its variants stand out for their impressive performance, which is widely attributed to their core mechanism of matching the student's output distribution to that of a pre-trained teacher model. In this work, we challenge this conventional understanding. Through a rigorous decomposition of the DMD training objective, we reveal that in complex tasks like text-to-image generation, where CFG is typically required for desirable few-step performance, the primary driver of few-step distillation is not distribution matching, but a previously overlooked component we identify as CFG Augmentation (CA). We demonstrate that this term acts as the core ``engine'' of distillation, while the Distribution Matching (DM) term functions as a ``regularizer'' that ensures training stability and mitigates artifacts. We further validate this decoupling by demonstrating that while the DM term is a highly effective regularizer, it is not unique; simpler non-parametric constraints or GAN-based objectives can serve the same stabilizing function, albeit with different trade-offs. This decoupling of labor motivates a more principled analysis of the properties of both terms, leading to a more systematic and in-depth understanding. This new understanding further enables us to propose principled modifications to the distillation process, such as decoupling the noise schedules for the engine and the regularizer, leading to further performance gains. Notably, our method has been adopted by the Z-Image ( https://github.com/Tongyi-MAI/Z-Image ) project to develop a top-tier 8-step image generation model, empirically validating the generalization and robustness of our findings.

Songpengcheng Xia, Yu Zhang, Zhuo Su et al.

Estimating full-body motion using the tracking signals of head and hands from VR devices holds great potential for various applications. However, the sparsity and unique distribution of observations present a significant challenge, resulting in an ill-posed problem with multiple feasible solutions (i.e., hypotheses). This amplifies uncertainty and ambiguity in full-body motion estimation, especially for the lower-body joints. Therefore, we propose a new method, EnvPoser, that employs a two-stage framework to perform full-body motion estimation using sparse tracking signals and pre-scanned environment from VR devices. EnvPoser models the multi-hypothesis nature of human motion through an uncertainty-aware estimation module in the first stage. In the second stage, we refine these multi-hypothesis estimates by integrating semantic and geometric environmental constraints, ensuring that the final motion estimation aligns realistically with both the environmental context and physical interactions. Qualitative and quantitative experiments on two public datasets demonstrate that our method achieves state-of-the-art performance, highlighting significant improvements in human motion estimation within motion-environment interaction scenarios.

Wentao Sun, Hanqing Xu, Quanyun Wu et al.

We introduce PointGauss, a novel point cloud-guided framework for real-time multi-object segmentation in Gaussian Splatting representations. Unlike existing methods that suffer from prolonged initialization and limited multi-view consistency, our approach achieves efficient 3D segmentation by directly parsing Gaussian primitives through a point cloud segmentation-driven pipeline. The key innovation lies in two aspects: (1) a point cloud-based Gaussian primitive decoder that generates 3D instance masks within 1 minute, and (2) a GPU-accelerated 2D mask rendering system that ensures multi-view consistency. Extensive experiments demonstrate significant improvements over previous state-of-the-art methods, achieving performance gains of 1.89 to 31.78% in multi-view mIoU, while maintaining superior computational efficiency. To address the limitations of current benchmarks (single-object focus, inconsistent 3D evaluation, small scale, and partial coverage), we present DesktopObjects-360, a novel comprehensive dataset for 3D segmentation in radiance fields, featuring: (1) complex multi-object scenes, (2) globally consistent 2D annotations, (3) large-scale training data (over 27 thousand 2D masks), (4) full 360° coverage, and (5) 3D evaluation masks.

Jianbo Gao, Keke Gai, Jing Yu et al.

Recent advancement in large-scale Artificial Intelligence (AI) models offering multimodal services have become foundational in AI systems, making them prime targets for model theft. Existing methods select Out-of-Distribution (OoD) data as backdoor watermarks and retrain the original model for copyright protection. However, existing methods are susceptible to malicious detection and forgery by adversaries, resulting in watermark evasion. In this work, we propose Model-\underline{ag}nostic Black-box Backdoor W\underline{ate}rmarking Framework (AGATE) to address stealthiness and robustness challenges in multimodal model copyright protection. Specifically, we propose an adversarial trigger generation method to generate stealthy adversarial triggers from ordinary dataset, providing visual fidelity while inducing semantic shifts. To alleviate the issue of anomaly detection among model outputs, we propose a post-transform module to correct the model output by narrowing the distance between adversarial trigger image embedding and text embedding. Subsequently, a two-phase watermark verification is proposed to judge whether the current model infringes by comparing the two results with and without the transform module. Consequently, we consistently outperform state-of-the-art methods across five datasets in the downstream tasks of multimodal image-text retrieval and image classification. Additionally, we validated the robustness of AGATE under two adversarial attack scenarios.

Qi Wu, Yingguang Yang, hao liu et al.

Social bot detection is crucial for mitigating misinformation, online manipulation, and coordinated inauthentic behavior. While existing neural network-based detectors perform well on benchmarks, they struggle with generalization due to distribution shifts across datasets and frequently produce overconfident predictions for out-of-distribution accounts beyond the training data. To address this, we introduce a novel Uncertainty Estimation for Social Bot Detection (UESBD) framework, which quantifies the predictive uncertainty of detectors beyond mere classification. For this task, we propose Robust Multi-modal Neural Processes (RMNP), which aims to enhance the robustness of multi-modal neural processes to modality inconsistencies caused by social bot camouflage. RMNP first learns unimodal representations through modality-specific encoders. Then, unimodal attentive neural processes are employed to encode the Gaussian distribution of unimodal latent variables. Furthermore, to avoid social bots stealing human features to camouflage themselves thus causing certain modalities to provide conflictive information, we introduce an evidential gating network to explicitly model the reliability of modalities. The joint latent distribution is learned through the generalized product of experts, which takes the reliability of each modality into consideration during fusion. The final prediction is obtained through Monte Carlo sampling of the joint latent distribution followed by a decoder. Experiments on three real-world benchmarks show the effectiveness of RMNP in classification and uncertainty estimation, as well as its robustness to modality conflicts.

Kaichen Xu, Qilong Wu, Yan Lu et al.

The detection of anomalous tissue regions (ATRs) within affected tissues is crucial in clinical diagnosis and pathological studies. Conventional automated ATR detection methods, primarily based on histology images alone, falter in cases where ATRs and normal tissues have subtle visual differences. The recent spatial transcriptomics (ST) technology profiles gene expressions across tissue regions, offering a molecular perspective for detecting ATRs. However, there is a dearth of ATR detection methods that effectively harness complementary information from both histology images and ST. To address this gap, we propose MEATRD, a novel ATR detection method that integrates histology image and ST data. MEATRD is trained to reconstruct image patches and gene expression profiles of normal tissue spots (inliers) from their multimodal embeddings, followed by learning a one-class classification AD model based on latent multimodal reconstruction errors. This strategy harmonizes the strengths of reconstruction-based and one-class classification approaches. At the heart of MEATRD is an innovative masked graph dual-attention transformer (MGDAT) network, which not only facilitates cross-modality and cross-node information sharing but also addresses the model over-generalization issue commonly seen in reconstruction-based AD methods. Additionally, we demonstrate that modality-specific, task-relevant information is collated and condensed in multimodal bottleneck encoding generated in MGDAT, marking the first theoretical analysis of the informational properties of multimodal bottleneck encoding. Extensive evaluations across eight real ST datasets reveal MEATRD's superior performance in ATR detection, surpassing various state-of-the-art AD methods. Remarkably, MEATRD also proves adept at discerning ATRs that only show slight visual deviations from normal tissues.

Yongqiang Yu, Xuhui Li, Hazza Mahmood et al.

Real-world deployment of Vision-and-Language Navigation (VLN) agents is constrained by the scarcity of reliable supervision after offline training. While recent adaptation methods attempt to mitigate distribution shifts via environment-driven self-supervision (e.g., entropy minimization), these signals are often noisy and can cause the agent to amplify its own mistakes during long-horizon sequential decision-making. In this paper, we propose a paradigm shift that positions user feedback, specifically episode-level success confirmations and goal-level corrections, as a primary and general-purpose supervision signal for VLN. Unlike internal confidence scores, user feedback is intent-aligned and in-situ consistent, directly correcting the agent's decoupling from user instructions. To effectively leverage this supervision, we introduce a user-feedback-driven learning framework featuring a topology-aware trajectory construction pipeline. This mechanism lifts sparse, goal-level corrections into dense path-level supervision by generating feasible paths on the agent's incrementally built topological graph, enabling sample-efficient imitation learning without requiring step-by-step human demonstrations. Furthermore, we develop a persistent memory bank mechanism for warm-start initialization, supporting the reuse of previously acquired topology and cached representations across navigation sessions. Extensive experiments on the GSA-R2R benchmark demonstrate that our approach transforms sparse interaction into robust supervision, consistently outperforming environment-driven baselines while exhibiting strong adaptability across diverse instruction styles.

Dengyang Jiang, Dongyang Liu, Zanyi Wang et al.

Distribution Matching Distillation (DMD) distills a pre-trained multi-step diffusion model to a few-step one to improve inference efficiency. However, the performance of the latter is often capped by the former. To circumvent this dilemma, we propose DMDR, a novel framework that combines Reinforcement Learning (RL) techniques into the distillation process. We show that for the RL of the few-step generator, the DMD loss itself is a more effective regularization compared to the traditional ones. In turn, RL can help to guide the mode coverage process in DMD more effectively. These allow us to unlock the capacity of the few-step generator by conducting distillation and RL simultaneously. Meanwhile, we design the dynamic distribution guidance and dynamic renoise sampling training strategies to improve the initial distillation process. The experiments demonstrate that DMDR can achieve leading visual quality, prompt coherence among few-step methods, and even exhibit performance that exceeds the multi-step teacher.

Yingguang Yang, Xianghua Zeng, Qi Wu et al.

Social networks have become a crucial source of real-time information for individuals. The influence of social bots within these platforms has garnered considerable attention from researchers, leading to the development of numerous detection technologies. However, the vulnerability and robustness of these detection methods is still underexplored. Existing Graph Neural Network (GNN)-based methods cannot be directly applied due to the issues of limited control over social agents, the black-box nature of bot detectors, and the heterogeneity of bots. To address these challenges, this paper proposes the first adversarial multi-agent Reinforcement learning framework for social Bot control attacks (RoBCtrl) targeting GNN-based social bot detectors. Specifically, we use a diffusion model to generate high-fidelity bot accounts by reconstructing existing account data with minor modifications, thereby evading detection on social platforms. To the best of our knowledge, this is the first application of diffusion models to mimic the behavior of evolving social bots effectively. We then employ a Multi-Agent Reinforcement Learning (MARL) method to simulate bots adversarial behavior. We categorize social accounts based on their influence and budget. Different agents are then employed to control bot accounts across various categories, optimizing the attachment strategy through reinforcement learning. Additionally, a hierarchical state abstraction based on structural entropy is designed to accelerate the reinforcement learning. Extensive experiments on social bot detection datasets demonstrate that our framework can effectively undermine the performance of GNN-based detectors.

Wentao Sun, Quanyun Wu, Hanqing Xu et al.

3D Gaussian Splatting (3DGS) has emerged as a powerful paradigm for explicit 3D scene representation, yet achieving efficient and consistent 3D segmentation remains challenging. Current methods suffer from prohibitive computational costs, limited 3D spatial reasoning, and an inability to track multiple objects simultaneously. We present Segment Any Gaussians Online (SAGOnline), a lightweight and zero-shot framework for real-time 3D segmentation in Gaussian scenes that addresses these limitations through two key innovations: (1) a decoupled strategy that integrates video foundation models (e.g., SAM2) for view-consistent 2D mask propagation across synthesized views; and (2) a GPU-accelerated 3D mask generation and Gaussian-level instance labeling algorithm that assigns unique identifiers to 3D primitives, enabling lossless multi-object tracking and segmentation across views. SAGOnline achieves state-of-the-art performance on NVOS (92.7% mIoU) and Spin-NeRF (95.2% mIoU) benchmarks, outperforming Feature3DGS, OmniSeg3D-gs, and SA3D by 15--1500 times in inference speed (27 ms/frame). Qualitative results demonstrate robust multi-object segmentation and tracking in complex scenes. Our contributions include: (i) a lightweight and zero-shot framework for 3D segmentation in Gaussian scenes, (ii) explicit labeling of Gaussian primitives enabling simultaneous segmentation and tracking, and (iii) the effective adaptation of 2D video foundation models to the 3D domain. This work allows real-time rendering and 3D scene understanding, paving the way for practical AR/VR and robotic applications.

Wenfeng Lin, Renjie Chen, Boyuan Liu et al.

Recent advances in video generation demand increasingly efficient training recipes to mitigate escalating computational costs. In this report, we present ContentV, an 8B-parameter text-to-video model that achieves state-of-the-art performance (85.14 on VBench) after training on 256 x 64GB Neural Processing Units (NPUs) for merely four weeks. ContentV generates diverse, high-quality videos across multiple resolutions and durations from text prompts, enabled by three key innovations: (1) A minimalist architecture that maximizes reuse of pre-trained image generation models for video generation; (2) A systematic multi-stage training strategy leveraging flow matching for enhanced efficiency; and (3) A cost-effective reinforcement learning with human feedback framework that improves generation quality without requiring additional human annotations. All the code and models are available at: https://contentv.github.io.

Changxun Zhu, Qilong Wu, Lingjuan Lyu et al.

Federated learning is vulnerable to poisoning attacks by malicious adversaries. Existing methods often involve high costs to achieve effective attacks. To address this challenge, we propose a sybil-based virtual data poisoning attack, where a malicious client generates sybil nodes to amplify the poisoning model's impact. To reduce neural network computational complexity, we develop a virtual data generation method based on gradient matching. We also design three schemes for target model acquisition, applicable to online local, online global, and offline scenarios. In simulation, our method outperforms other attack algorithms since our method can obtain a global target model under non-independent uniformly distributed data.

Liangqi Lei, Keke Gai, Jing Yu et al.

Latent diffusion models have exhibited considerable potential in generative tasks. Watermarking is considered to be an alternative to safeguard the copyright of generative models and prevent their misuse. However, in the context of model distribution scenarios, the accessibility of models to large scale of model users brings new challenges to the security, efficiency and robustness of existing watermark solutions. To address these issues, we propose a secure and efficient watermarking solution. A new security mechanism is designed to prevent watermark leakage and watermark escape, which considers watermark randomness and watermark-model association as two constraints for mandatory watermark injection. To reduce the time cost of training the security module, watermark injection and the security mechanism are decoupled, ensuring that fine-tuning VAE only accomplishes the security mechanism without the burden of learning watermark patterns. A watermark distribution-based verification strategy is proposed to enhance the robustness against diverse attacks in the model distribution scenarios. Experimental results prove that our watermarking consistently outperforms existing six baselines on effectiveness and robustness against ten image processing attacks and adversarial attacks, while enhancing security in the distribution scenarios.

Salim Rezvani, Farhad Pourpanah, Chee Peng Lim et al.

This paper presents a review on methods for class-imbalanced learning with the Support Vector Machine (SVM) and its variants. We first explain the structure of SVM and its variants and discuss their inefficiency in learning with class-imbalanced data sets. We introduce a hierarchical categorization of SVM-based models with respect to class-imbalanced learning. Specifically, we categorize SVM-based models into re-sampling, algorithmic, and fusion methods, and discuss the principles of the representative models in each category. In addition, we conduct a series of empirical evaluations to compare the performances of various representative SVM-based models in each category using benchmark imbalanced data sets, ranging from low to high imbalanced ratios. Our findings reveal that while algorithmic methods are less time-consuming owing to no data pre-processing requirements, fusion methods, which combine both re-sampling and algorithmic approaches, generally perform the best, but with a higher computational load. A discussion on research gaps and future research directions is provided.

Farhad Pourpanah, Chee Peng Lim, Ali Etemad et al.

Most semi-supervised learning (SSL) models entail complex structures and iterative training processes as well as face difficulties in interpreting their predictions to users. To address these issues, this paper proposes a new interpretable SSL model using the supervised and unsupervised Adaptive Resonance Theory (ART) family of networks, which is denoted as SSL-ART. Firstly, SSL-ART adopts an unsupervised fuzzy ART network to create a number of prototype nodes using unlabeled samples. Then, it leverages a supervised fuzzy ARTMAP structure to map the established prototype nodes to the target classes using labeled samples. Specifically, a one-to-many (OtM) mapping scheme is devised to associate a prototype node with more than one class label. The main advantages of SSL-ART include the capability of: (i) performing online learning, (ii) reducing the number of redundant prototype nodes through the OtM mapping scheme and minimizing the effects of noisy samples, and (iii) providing an explanation facility for users to interpret the predicted outcomes. In addition, a weighted voting strategy is introduced to form an ensemble SSL-ART model, which is denoted as WESSL-ART. Every ensemble member, i.e., SSL-ART, assigns {\color{black}a different weight} to each class based on its performance pertaining to the corresponding class. The aim is to mitigate the effects of training data sequences on all SSL-ART members and improve the overall performance of WESSL-ART. The experimental results on eighteen benchmark data sets, three artificially generated data sets, and a real-world case study indicate the benefits of the proposed SSL-ART and WESSL-ART models for tackling pattern classification problems.

Wenyan Pan, Zhili Zhou, Miaogen Ling et al.

The objective of image manipulation detection is to identify and locate the manipulated regions in the images. Recent approaches mostly adopt the sophisticated Convolutional Neural Networks (CNNs) to capture the tampering artifacts left in the images to locate the manipulated regions. However, these approaches ignore the feature correlations, i.e., feature inconsistencies, between manipulated regions and non-manipulated regions, leading to inferior detection performance. To address this issue, we propose a hierarchical Graph Convolutional Network (HGCN-Net), which consists of two parallel branches: the backbone network branch and the hierarchical graph representation learning (HGRL) branch for image manipulation detection. Specifically, the feature maps of a given image are extracted by the backbone network branch, and then the feature correlations within the feature maps are modeled as a set of fully-connected graphs for learning the hierarchical graph representation by the HGRL branch. The learned hierarchical graph representation can sufficiently capture the feature correlations across different scales, and thus it provides high discriminability for distinguishing manipulated and non-manipulated regions. Extensive experiments on four public datasets demonstrate that the proposed HGCN-Net not only provides promising detection accuracy, but also achieves strong robustness under a variety of common image attacks in the task of image manipulation detection, compared to the state-of-the-arts.

Yurong Chen, Hui Zhang, Yaonan Wang et al.

Anomaly detection (AD) has been an active research area in various domains. Yet, the increasing data scale, complexity, and dimension turn the traditional methods into challenging. Recently, the deep generative model, such as the variational autoencoder (VAE), has sparked a renewed interest in the AD problem. However, the probability distribution divergence used as the regularization is too strong, which causes the model cannot capture the manifold of the true data. In this paper, we propose the Projected Sliced Wasserstein (PSW) autoencoder-based anomaly detection method. Rooted in the optimal transportation, the PSW distance is a weaker distribution measure compared with $f$-divergence. In particular, the computation-friendly eigen-decomposition method is leveraged to find the principal component for slicing the high-dimensional data. In this case, the Wasserstein distance can be calculated with the closed-form, even the prior distribution is not Gaussian. Comprehensive experiments conducted on various real-world hyperspectral anomaly detection benchmarks demonstrate the superior performance of the proposed method.

Yimin Yang, Wandong Zhang, Jonathan Wu et al.

2D Convolutional neural network (CNN) has arguably become the de facto standard for computer vision tasks. Recent findings, however, suggest that CNN may not be the best option for 1D pattern recognition, especially for datasets with over 1 M training samples, e.g., existing CNN-based methods for 1D signals are highly reliant on human pre-processing. Common practices include utilizing discrete Fourier transform (DFT) to reconstruct 1D signal into 2D array. To add to extant knowledge, in this paper, a novel 1D data processing algorithm is proposed for 1D big data analysis through learning a deep deconvolutional-convolutional network. Rather than resorting to human-based techniques, we employed deconvolution layers to convert 1 D signals into 2D data. On top of the deconvolution model, the data was identified by a 2D CNN. Compared with the existing 1D signal processing algorithms, DCNet boasts the advantages of less human-made inference and higher generalization performance. Our experimental results from a varying number of training patterns (50 K to 11 M) from classification and regression demonstrate the desirability of our new approach.

Wandong Zhang, QM Jonathan Wu, Yimin Yang et al.

Most multilayer least squares (LS)-based neural networks are structured with two separate stages: unsupervised feature encoding and supervised pattern classification. Once the unsupervised learning is finished, the latent encoding would be fixed without supervised fine-tuning. However, in complex tasks such as handling the ImageNet dataset, there are often many more clues that can be directly encoded, while the unsupervised learning, by definition cannot know exactly what is useful for a certain task. This serves as the motivation to retrain the latent space representations to learn some clues that unsupervised learning has not yet learned. In particular, the error matrix from the output layer is pulled back to each hidden layer, and the parameters of the hidden layer are recalculated with Moore-Penrose (MP) inverse for more generalized representations. In this paper, a recomputation-based multilayer network using MP inverse (RML-MP) is developed. A sparse RML-MP (SRML-MP) model to boost the performance of RML-MP is then proposed. The experimental results with varying training samples (from 3 K to 1.8 M) show that the proposed models provide better generalization performance than most representation learning algorithms.

Wandong Zhang, Yimin Yang, Jonathan Wu

Recent work [1] has utilized Moore-Penrose (MP) inverse in deep convolutional neural network (DCNN) learning, which achieves better generalization performance over the DCNN with a stochastic gradient descent (SGD) pipeline. However, Yang's work has not gained much popularity in practice due to its high sensitivity of hyper-parameters and stringent demands of computational resources. To enhance its applicability, this paper proposes a novel MP inverse-based fast retraining strategy. In each training epoch, a random learning strategy that controls the number of convolutional layers trained in the backward pass is first utilized. Then, an MP inverse-based batch-by-batch learning strategy, which enables the network to be implemented without access to industrial-scale computational resources, is developed to refine the dense layer parameters. Experimental results empirically demonstrate that fast retraining is a unified strategy that can be used for all DCNNs. Compared to other learning strategies, the proposed learning pipeline has robustness against the hyper-parameters, and the requirement of computational resources is significantly reduced. [1] Y. Yang, J. Wu, X. Feng, and A. Thangarajah, "Recomputation of dense layers for the perfor-238mance improvement of dcnn," IEEE Trans. Pattern Anal. Mach. Intell., 2019.

Zhilin Yu, Chao Wang, Xin Wang et al.

Modern deep neural networks rely on overparameterization to achieve state-of-the-art generalization. But overparameterized models are computationally expensive. Network pruning is often employed to obtain less demanding models for deployment. Fine-grained pruning removes individual weights in parameter tensors and can achieve a high model compression ratio with little accuracy degradation. However, it introduces irregularity into the computing dataflow and often does not yield improved model inference efficiency in practice. Coarse-grained model pruning, while realizing satisfactory inference speedup through removal of network weights in groups, e.g. an entire filter, often lead to significant accuracy degradation. This work introduces the cross-channel intragroup (CCI) sparsity structure, which can prevent the inference inefficiency of fine-grained pruning while maintaining outstanding model performance. We then present a novel training algorithm designed to perform well under the constraint imposed by the CCI-Sparsity. Through a series of comparative experiments we show that our proposed CCI-Sparsity structure and the corresponding pruning algorithm outperform prior art in inference efficiency by a substantial margin given suited hardware acceleration in the future.

Yimin Yang, Q. M. Jonathan Wu, Xiexing Feng et al.

An iterative method of learning has become a paradigm for training deep convolutional neural networks (DCNN). However, utilizing a non-iterative learning strategy can accelerate the training process of the DCNN and surprisingly such approach has been rarely explored by the deep learning (DL) community. It motivates this paper to introduce a non-iterative learning strategy that eliminates the backpropagation (BP) at the top dense or fully connected (FC) layers of DCNN, resulting in, lower training time and higher performance. The proposed method exploits the Moore-Penrose Inverse to pull back the current residual error to each FC layer, generating well-generalized features. Then using the recomputed features, i.e., the new generalized features the weights of each FC layer is computed according to the Moore-Penrose Inverse. We evaluate the proposed approach on six widely accepted object recognition benchmark datasets: Scene-15, CIFAR-10, CIFAR-100, SUN-397, Places365, and ImageNet. The experimental results show that the proposed method obtains significant improvements over 30 state-of-the-art methods. Interestingly, it also indicates that any DCNN with the proposed method can provide better performance than the same network with its original training based on BP.

Thangarajah Akilan, Q. M. Jonathan Wu, Wei Jiang

Following the rapidly growing digital image usage, automatic image categorization has become preeminent research area. It has broaden and adopted many algorithms from time to time, whereby multi-feature (generally, hand-engineered features) based image characterization comes handy to improve accuracy. Recently, in machine learning, pre-trained deep convolutional neural networks (DCNNs or ConvNets) have been that the features extracted through such DCNN can improve classification accuracy. Thence, in this paper, we further investigate a feature embedding strategy to exploit cues from multiple DCNNs. We derive a generalized feature space by embedding three different DCNN bottleneck features with weights respect to their Softmax cross-entropy loss. Test outcomes on six different object classification data-sets and an action classification data-set show that regardless of variation in image statistics and tasks the proposed multi-DCNN bottleneck feature fusion is well suited to image classification tasks and an effective complement of DCNN. The comparisons to existing fusion-based image classification approaches prove that the proposed method surmounts the state-of-the-art methods and produces competitive results with fully trained DCNNs as well.

Jiawei Chen, Jonathan Wu, Janusz Konrad et al.

Deep convolutional neural networks (ConvNets) have been recently shown to attain state-of-the-art performance for action recognition on standard-resolution videos. However, less attention has been paid to recognition performance at extremely low resolutions (eLR) (e.g., 16 x 12 pixels). Reliable action recognition using eLR cameras would address privacy concerns in various application environments such as private homes, hospitals, nursing/rehabilitation facilities, etc. In this paper, we propose a semi-coupled filter-sharing network that leverages high resolution (HR) videos during training in order to assist an eLR ConvNet. We also study methods for fusing spatial and temporal ConvNets customized for eLR videos in order to take advantage of appearance and motion information. Our method outperforms state-of-the-art methods at extremely low resolutions on IXMAS (93.7%) and HMDB (29.2%) datasets.

Ashirbani Saha, Q. M. Jonathan Wu

A blind approach to evaluate the perceptual sharpness present in a natural image is proposed. Though the literature demonstrates a set of variegated visual cues to detect or evaluate the absence or presence of sharpness, we emphasize in the current work that high frequency content and local standard deviation can form strong features to compute perceived sharpness in any natural image, and can be considered an able alternative for the existing cues. Unsharp areas in a natural image happen to exhibit uniform intensity or lack of sharp changes between regions. Sharp region transitions in an image are caused by the presence of spatial high frequency content. Therefore, in the proposed approach, we hypothesize that using the high frequency content as the principal stimulus, the perceived sharpness can be quantified in an image. When an image is convolved with a high pass filter, higher values at any pixel location signify the presence of high frequency content at those locations. Considering these values as the stimulus, the exponent of the stimulus is weighted by local standard deviation to impart the contribution of the local contrast within the formation of the sharpness map. The sharpness map highlights the relatively sharper regions in the image and is used to calculate the perceived sharpness score of the image. The advantages of the proposed method lie in its use of simple visual cues of high frequency content and local contrast to arrive at the perceptual score, and requiring no training with the images. The promise of the proposed method is demonstrated by its ability to compute perceived sharpness for within image and across image sharpness changes and for blind evaluation of perceptual degradation resulting due to presence of blur. Experiments conducted on several databases demonstrate improved performance of the proposed method over that of the state-of-the-art techniques.

Ashirbani Saha, Q. M. Jonathan Wu

Full-reference image quality assessment (FR-IQA) techniques compare a reference and a distorted/test image and predict the perceptual quality of the test image in terms of a scalar value representing an objective score. The evaluation of FR-IQA techniques is carried out by comparing the objective scores from the techniques with the subjective scores (obtained from human observers) provided in the image databases used for the IQA. Hence, we reasonably assume that the goal of a human observer is to rate the distortion present in the test image. The goal oriented tasks are processed by the human visual system (HVS) through top-down processing which actively searches for local distortions driven by the goal. Therefore local distortion measures in an image are important for the top-down processing. At the same time, bottom-up processing also takes place signifying spontaneous visual functions in the HVS. To account for this, global perceptual features can be used. Therefore, we hypothesize that the resulting objective score for an image can be derived from the combination of local and global distortion measures calculated from the reference and test images. We calculate the local distortion by measuring the local correlation differences from the gradient and contrast information. For global distortion, dissimilarity of the saliency maps computed from a bottom-up model of saliency is used. The motivation behind the proposed approach has been thoroughly discussed, accompanied by an intuitive analysis. Finally, experiments are conducted in six benchmark databases suggesting the effectiveness of the proposed approach that achieves competitive performance with the state-of-the-art methods providing an improvement in the overall performance.

Yimin Yang, Q. M. Jonathan Wu, Guangbin Huang et al.

According to conventional neural network theories, the feature of single-hidden-layer feedforward neural networks(SLFNs) resorts to parameters of the weighted connections and hidden nodes. SLFNs are universal approximators when at least the parameters of the networks including hidden-node parameter and output weight are exist. Unlike above neural network theories, this paper indicates that in order to let SLFNs work as universal approximators, one may simply calculate the hidden node parameter only and the output weight is not needed at all. In other words, this proposed neural network architecture can be considered as a standard SLFNs with fixing output weight equal to an unit vector. Further more, this paper presents experiments which show that the proposed learning method tends to extremely reduce network output error to a very small number with only 1 hidden node. Simulation results demonstrate that the proposed method can provide several to thousands of times faster than other learning algorithm including BP, SVM/SVR and other ELM methods.