Zhiyong Zhang

Jie Feng, Ruimin Feng, Qing Wu et al.

Supervised Deep-Learning (DL)-based reconstruction algorithms have shown state-of-the-art results for highly-undersampled dynamic Magnetic Resonance Imaging (MRI) reconstruction. However, the requirement of excessive high-quality ground-truth data hinders their applications due to the generalization problem. Recently, Implicit Neural Representation (INR) has appeared as a powerful DL-based tool for solving the inverse problem by characterizing the attributes of a signal as a continuous function of corresponding coordinates in an unsupervised manner. In this work, we proposed an INR-based method to improve dynamic MRI reconstruction from highly undersampled k-space data, which only takes spatiotemporal coordinates as inputs. Specifically, the proposed INR represents the dynamic MRI images as an implicit function and encodes them into neural networks. The weights of the network are learned from sparsely-acquired (k, t)-space data itself only, without external training datasets or prior images. Benefiting from the strong implicit continuity regularization of INR together with explicit regularization for low-rankness and sparsity, our proposed method outperforms the compared scan-specific methods at various acceleration factors. E.g., experiments on retrospective cardiac cine datasets show an improvement of 5.5 ~ 7.1 dB in PSNR for extremely high accelerations (up to 41.6-fold). The high-quality and inner continuity of the images provided by INR has great potential to further improve the spatiotemporal resolution of dynamic MRI, without the need of any training data.

YuYang Sun, ZhiYong Zhang, Isao Echizen et al.

Deep-learning-based technologies such as deepfakes ones have been attracting widespread attention in both society and academia, particularly ones used to synthesize forged face images. These automatic and professional-skill-free face manipulation technologies can be used to replace the face in an original image or video with any target object while maintaining the expression and demeanor. Since human faces are closely related to identity characteristics, maliciously disseminated identity manipulated videos could trigger a crisis of public trust in the media and could even have serious political, social, and legal implications. To effectively detect manipulated videos, we focus on the position offset in the face blending process, resulting from the forced affine transformation of the normalized forged face. We introduce a method for detecting manipulated videos that is based on the trajectory of the facial region displacement. Specifically, we develop a virtual-anchor-based method for extracting the facial trajectory, which can robustly represent displacement information. This information was used to construct a network for exposing multidimensional artifacts in the trajectory sequences of manipulated videos that is based on dual-stream spatial-temporal graph attention and a gated recurrent unit backbone. Testing of our method on various manipulation datasets demonstrated that its accuracy and generalization ability is competitive with that of the leading detection methods.

Zhiyong Zhang, Pushyami Kaveti, Hanumant Singh et al.

In this paper, we present a methodology for fisheries-related data that allows us to converge on a labeled image dataset by iterating over the dataset with multiple training and production loops that can exploit crowdsourcing interfaces. We present our algorithm and its results on two separate sets of image data collected using the Seabed autonomous underwater vehicle. The first dataset comprises of 2,026 completely unlabeled images, while the second consists of 21,968 images that were point annotated by experts. Our results indicate that training with a small subset and iterating on that to build a larger set of labeled data allows us to converge to a fully annotated dataset with a small number of iterations. Even in the case of a dataset labeled by experts, a single iteration of the methodology improves the labels by discovering additional complicated examples of labels associated with fish that overlap, are very small, or obscured by the contrast limitations associated with underwater imagery.

Zhiyong Zhang, Aniket Gupta, Huaizu Jiang et al.

Real-time high-accuracy optical flow estimation is critical for a variety of real-world robotic applications. However, current learning-based methods often struggle to balance accuracy and computational efficiency: methods that achieve high accuracy typically demand substantial processing power, while faster approaches tend to sacrifice precision. These fast approaches specifically falter in their generalization capabilities and do not perform well across diverse real-world scenarios. In this work, we revisit the limitations of the SOTA methods and present NeuFlow-V2, a novel method that offers both - high accuracy in real-world datasets coupled with low computational overhead. In particular, we introduce a novel light-weight backbone and a fast refinement module to keep computational demands tractable while delivering accurate optical flow. Experimental results on synthetic and real-world datasets demonstrate that NeuFlow-V2 provides similar accuracy to SOTA methods while achieving 10x-70x speedups. It is capable of running at over 20 FPS on 512x384 resolution images on a Jetson Orin Nano. The full training and evaluation code is available at https://github.com/neufieldrobotics/NeuFlow_v2.

Xin Yuan, Zhiyong Zhang, Xin Xu et al.

With the increasing demand for robust person Re-ID in unconstrained environments, learning from datasets with noisy labels and sparse per-identity samples remains a critical challenge. Existing noise-robust person Re-ID methods primarily rely on loss-correction or sample-selection strategies using softmax outputs. However, these methods suffer from two key limitations: 1) Softmax exhibits translation invariance, leading to over-confident and unreliable predictions on corrupted labels. 2) Conventional sample selection based on small-loss criteria often discards valuable hard positives that are crucial for learning discriminative features. To overcome these issues, we propose the CAlibration-to-REfinement (CARE) method, a two-stage framework that seeks certainty through probabilistic evidence propagation from calibration to refinement. In the calibration stage, we propose the probabilistic evidence calibration (PEC) that dismantles softmax translation invariance by injecting adaptive learnable parameters into the similarity function, and employs an evidential calibration loss to mitigate overconfidence on mislabeled samples. In the refinement stage, we design the evidence propagation refinement (EPR) that can more accurately distinguish between clean and noisy samples. Specifically, the EPR contains two steps: Firstly, the composite angular margin (CAM) metric is proposed to precisely distinguish clean but hard-to-learn positive samples from mislabeled ones in a hyperspherical space; Secondly, the certainty-oriented sphere weighting (COSW) is developed to dynamically allocate the importance of samples according to CAM, ensuring clean instances drive model updates. Extensive experimental results on Market1501, DukeMTMC-ReID, and CUHK03 datasets under both random and patterned noises show that CARE achieves competitive performance.

Xulong Zhang, Jianzong Wang, Ning Cheng et al.

The recent emergence of joint CTC-Attention model shows significant improvement in automatic speech recognition (ASR). The improvement largely lies in the modeling of linguistic information by decoder. The decoder joint-optimized with an acoustic encoder renders the language model from ground-truth sequences in an auto-regressive manner during training. However, the training corpus of the decoder is limited to the speech transcriptions, which is far less than the corpus needed to train an acceptable language model. This leads to poor robustness of decoder. To alleviate this problem, we propose linguistic-enhanced transformer, which introduces refined CTC information to decoder during training process, so that the decoder can be more robust. Our experiments on AISHELL-1 speech corpus show that the character error rate (CER) is relatively reduced by up to 7%. We also find that in joint CTC-Attention ASR model, decoder is more sensitive to linguistic information than acoustic information.

Yang Liu, Changzhen Qiu, Zhiyong Zhang

3D human pose estimation and mesh recovery have attracted widespread research interest in many areas, such as computer vision, autonomous driving, and robotics. Deep learning on 3D human pose estimation and mesh recovery has recently thrived, with numerous methods proposed to address different problems in this area. In this paper, to stimulate future research, we present a comprehensive review of recent progress over the past five years in deep learning methods for this area by delving into over 200 references. To the best of our knowledge, this survey is arguably the first to comprehensively cover deep learning methods for 3D human pose estimation, including both single-person and multi-person approaches, as well as human mesh recovery, encompassing methods based on explicit models and implicit representations. We also present comparative results on several publicly available datasets, together with insightful observations and inspiring future research directions. A regularly updated project page can be found at https://github.com/liuyangme/SOTA-3DHPE-HMR.

Zhiyong Zhang, Huaizu Jiang, Hanumant Singh

Real-time high-accuracy optical flow estimation is a crucial component in various applications, including localization and mapping in robotics, object tracking, and activity recognition in computer vision. While recent learning-based optical flow methods have achieved high accuracy, they often come with heavy computation costs. In this paper, we propose a highly efficient optical flow architecture, called NeuFlow, that addresses both high accuracy and computational cost concerns. The architecture follows a global-to-local scheme. Given the features of the input images extracted at different spatial resolutions, global matching is employed to estimate an initial optical flow on the 1/16 resolution, capturing large displacement, which is then refined on the 1/8 resolution with lightweight CNN layers for better accuracy. We evaluate our approach on Jetson Orin Nano and RTX 2080 to demonstrate efficiency improvements across different computing platforms. We achieve a notable 10x-80x speedup compared to several state-of-the-art methods, while maintaining comparable accuracy. Our approach achieves around 30 FPS on edge computing platforms, which represents a significant breakthrough in deploying complex computer vision tasks such as SLAM on small robots like drones. The full training and evaluation code is available at https://github.com/neufieldrobotics/NeuFlow.

Yuhan Xie, Kexin Jiang, Zhiyong Zhang et al.

Medical image segmentation based on deep learning is often faced with the problems of insufficient datasets and long time-consuming labeling. In this paper, we introduce the self-supervised method MAE(Masked Autoencoders) into knee joint images to provide a good initial weight for the segmentation model and improve the adaptability of the model to small datasets. Secondly, we propose a weakly supervised paradigm for meniscus segmentation based on the combination of point and line to reduce the time of labeling. Based on the weak label ,we design a region growing algorithm to generate pseudo-label. Finally we train the segmentation network based on pseudo-labels with weight transfer from self-supervision. Sufficient experimental results show that our proposed method combining self-supervision and weak supervision can almost approach the performance of purely fully supervised models while greatly reducing the required labeling time and dataset size.

Yuhan Xie, Zhiyong Zhang, Shaolong Chen et al.

The segmentation of atrial scan images is of great significance for the three-dimensional reconstruction of the atrium and the surgical positioning. Most of the existing segmentation networks adopt a 2D structure and only take original images as input, ignoring the context information of 3D images and the role of prior information. In this paper, we propose an atrium segmentation network LGSANet with location guidance and siamese adjustment, which takes adjacent three slices of images as input and adopts an end-to-end approach to achieve coarse-to-fine atrial segmentation. The location guidance(LG) block uses the prior information of the localization map to guide the encoding features of the fine segmentation stage, and the siamese adjustment(SA) block uses the context information to adjust the segmentation edges. On the atrium datasets of ACDC and ASC, sufficient experiments prove that our method can adapt to many classic 2D segmentation networks, so that it can obtain significant performance improvements.

Yang Liu, Zhiyong Zhang

The current methods of video-based 3D human pose estimation have achieved significant progress.However, they still face pressing challenges, such as the underutilization of spatiotemporal bodystructure features in transformers and the inadequate granularity of spatiotemporal interaction modeling in graph convolutional networks, which leads to pervasive depth ambiguity in monocular 3D human pose estimation. To address these limitations, this paper presents the Spatio-Temporal GraphFormer framework (STGFormer) for 3D human pose estimation in videos. First, we introduce a Spatio-Temporal criss-cross Graph (STG) attention mechanism, designed to more effectively leverage the inherent graph priors of the human body within continuous sequence distributions while capturing spatiotemporal long-range dependencies. Next, we present a dual-path Modulated Hop-wise Regular GCN (MHR-GCN) to independently process temporal and spatial dimensions in parallel, preserving features rich in temporal dynamics and the original or high-dimensional representations of spatial structures. Furthermore, the module leverages modulation to optimize parameter efficiency and incorporates spatiotemporal hop-wise skip connections to capture higher-order information. Finally, we demonstrate that our method achieves state-of-the-art performance on the Human3.6M and MPIINF-3DHP datasets.

Yang Liu, Zhiyong Zhang

3D human mesh recovery from monocular RGB images aims to estimate anatomically plausible 3D human models for downstream applications, but remains challenging under partial or severe occlusions. Regression-based methods are efficient yet often produce implausible or inaccurate results in unconstrained scenarios, while diffusion-based methods provide strong generative priors for occluded regions but may weaken fidelity to rare poses due to over-reliance on generation. To address these limitations, we propose a brain-inspired synergistic framework that integrates the discriminative power of vision transformers with the generative capability of conditional diffusion models. Specifically, the ViT-based pathway extracts deterministic visual cues from visible regions, while the diffusion-based pathway synthesizes structurally coherent human body representations. To effectively bridge the two pathways, we design a diverse-consistent feature learning module to align discriminative features with generative priors, and a cross-attention multi-level fusion mechanism to enable bidirectional interaction across semantic levels. Experiments on standard benchmarks demonstrate that our method achieves superior performance on key metrics and shows strong robustness in complex real-world scenarios.

Yang Liu, Zhiyong Zhang

Real-time, high-fidelity 3D human reconstruction from RGB images is essential for interactive applications such as virtual reality and gaming, yet remains challenging due to the complex non-rigid deformations of dynamic human bodies. Although 3D Gaussian Splatting enables efficient rendering, existing methods struggle to capture fine geometric details and often produce artifacts such as fused fingers and over-smoothed faces. Moreover, conventional spatial-field-based dynamic modeling faces a trade-off between reconstruction fidelity and GPU memory consumption. To address these issues, we propose a novel 3D Gaussian human reconstruction framework that combines region-aware initialization with rich geometric priors. Specifically, we leverage the expressive SMPL-X model to initialize both 3D Gaussians and skinning weights, providing a robust geometric foundation for precise reconstruction. We further introduce a region-aware density initialization strategy and a geometry-aware multi-scale hash encoding module to improve local detail recovery while maintaining computational efficiency.Experiments on PeopleSnapshot and GalaBasketball show that our method achieves superior reconstruction quality and finer detail preservation under complex motions, while maintaining real-time rendering speed.

Yuyang Sun, Huy H. Nguyen, Chun-Shien Lu et al.

The growing diversity of digital face manipulation techniques has led to an urgent need for a universal and robust detection technology to mitigate the risks posed by malicious forgeries. We present a blended-based detection approach that has robust applicability to unseen datasets. It combines a method for generating synthetic training samples, i.e., reconstructed blended images, that incorporate potential deepfake generator artifacts and a detection model, a multi-scale feature reconstruction network, for capturing the generic boundary artifacts and noise distribution anomalies brought about by digital face manipulations. Experiments demonstrated that this approach results in better performance in both cross-manipulation detection and cross-dataset detection on unseen data.

Congjing Yu, Jing Ye, Yang Liu et al.

Multimodal medical analysis combining image and tabular data has gained increasing attention. However, effective fusion remains challenging due to cross-modal discrepancies in feature dimensions and modality contributions, as well as the noise from high-dimensional tabular inputs. To address these problems, we present AMF-MedIT, an efficient Align-Modulation-Fusion framework for medical image and tabular data integration, particularly under data-scarce conditions. Built upon a self-supervised learning strategy, we introduce the Adaptive Modulation and Fusion (AMF) module, a novel, streamlined fusion paradigm that harmonizes dimension discrepancies and dynamically balances modality contributions. It integrates prior knowledge to guide the allocation of modality contributions in the fusion and employs feature masks together with magnitude and leakage losses to adjust the dimensionality and magnitude of unimodal features. Additionally, we develop FT-Mamba, a powerful tabular encoder leveraging a selective mechanism to handle noisy medical tabular data efficiently. Extensive experiments, including simulations of clinical noise, demonstrate that AMF-MedIT achieves superior accuracy, robustness, and data efficiency across multimodal classification tasks. Interpretability analyses further reveal how FT-Mamba shapes multimodal pretraining and enhances the image encoder's attention, highlighting the practical value of our framework for reliable and efficient clinical artificial intelligence applications.

Yuyang Sun, Zhiyong Zhang, Changzhen Qiu et al.

With the rapid development of generation model, AI-based face manipulation technology, which called DeepFakes, has become more and more realistic. This means of face forgery can attack any target, which poses a new threat to personal privacy and property security. Moreover, the misuse of synthetic video shows potential dangers in many areas, such as identity harassment, pornography and news rumors. Inspired by the fact that the spatial coherence and temporal consistency of physiological signal are destroyed in the generated content, we attempt to find inconsistent patterns that can distinguish between real videos and synthetic videos from the variations of facial pixels, which are highly related to physiological information. Our approach first applies Eulerian Video Magnification (EVM) at multiple Gaussian scales to the original video to enlarge the physiological variations caused by the change of facial blood volume, and then transform the original video and magnified videos into a Multi-Scale Eulerian Magnified Spatial-Temporal map (MEMSTmap), which can represent time-varying physiological enhancement sequences on different octaves. Then, these maps are reshaped into frame patches in column units and sent to the vision Transformer to learn the spatio-time descriptors of frame levels. Finally, we sort out the feature embedding and output the probability of judging whether the video is real or fake. We validate our method on the FaceForensics++ and DeepFake Detection datasets. The results show that our model achieves excellent performance in forgery detection, and also show outstanding generalization capability in cross-data domain.

Xueli Jia, Jianzong Wang, Zhiyong Zhang et al.

End-to-end Spoken Language Understanding (SLU) models are made increasingly large and complex to achieve the state-ofthe-art accuracy. However, the increased complexity of a model can also introduce high risk of over-fitting, which is a major challenge in SLU tasks due to the limitation of available data. In this paper, we propose an attention-based SLU model together with three encoder enhancement strategies to overcome data sparsity challenge. The first strategy focuses on the transferlearning approach to improve feature extraction capability of the encoder. It is implemented by pre-training the encoder component with a quantity of Automatic Speech Recognition annotated data relying on the standard Transformer architecture and then fine-tuning the SLU model with a small amount of target labelled data. The second strategy adopts multitask learning strategy, the SLU model integrates the speech recognition model by sharing the same underlying encoder, such that improving robustness and generalization ability. The third strategy, learning from Component Fusion (CF) idea, involves a Bidirectional Encoder Representation from Transformer (BERT) model and aims to boost the capability of the decoder with an auxiliary network. It hence reduces the risk of over-fitting and augments the ability of the underlying encoder, indirectly. Experiments on the FluentAI dataset show that cross-language transfer learning and multi-task strategies have been improved by up to 4:52% and 3:89% respectively, compared to the baseline.

Manzhang Xu, Bijun Tang, Chao Zhu et al.

Reducing the lateral scale of two-dimensional (2D) materials to one-dimensional (1D) has attracted substantial research interest not only to achieve competitive electronic device applications but also for the exploration of fundamental physical properties. Controllable synthesis of high-quality 1D nanoribbons (NRs) is thus highly desirable and essential for the further study. Traditional exploration of the optimal synthesis conditions of novel materials is based on the trial-and-error approach, which is time consuming, costly and laborious. Recently, machine learning (ML) has demonstrated promising capability in guiding material synthesis through effectively learning from the past data and then making recommendations. Here, we report the implementation of supervised ML for the chemical vapor deposition (CVD) synthesis of high-quality 1D few-layered WTe2 nanoribbons (NRs). The synthesis parameters of the WTe2 NRs are optimized by the trained ML model. On top of that, the growth mechanism of as-synthesized 1T' few-layered WTe2 NRs is further proposed, which may inspire the growth strategies for other 1D nanostructures. Our findings suggest that ML is a powerful and efficient approach to aid the synthesis of 1D nanostructures, opening up new opportunities for intelligent material development.

Chi Wang, Dong Liu, Yifeng Wei et al.

A software system has been developed for the DArk Matter Particle Explorer (DAMPE) mission, a satellite-based experiment. The DAMPE software is mainly written in C++ and steered using Python script. This article presents an overview of the DAMPE offline software, including the major architecture design and specific implementation for simulation, calibration and reconstruction. The whole system has been successfully applied to DAMPE data analysis, based on which some results from simulation and beam test experiments are obtained and presented.

Lantian Li, Yiye Lin, Zhiyong Zhang et al.

A deep learning approach has been proposed recently to derive speaker identifies (d-vector) by a deep neural network (DNN). This approach has been applied to text-dependent speaker recognition tasks and shows reasonable performance gains when combined with the conventional i-vector approach. Although promising, the existing d-vector implementation still can not compete with the i-vector baseline. This paper presents two improvements for the deep learning approach: a phonedependent DNN structure to normalize phone variation, and a new scoring approach based on dynamic time warping (DTW). Experiments on a text-dependent speaker recognition task demonstrated that the proposed methods can provide considerable performance improvement over the existing d-vector implementation.

Xi Ma, Xiaoxi Wang, Dong Wang et al.

Low-frequency words place a major challenge for automatic speech recognition (ASR). The probabilities of these words, which are often important name entities, are generally under-estimated by the language model (LM) due to their limited occurrences in the training data. Recently, we proposed a word-pair approach to deal with the problem, which borrows information of frequent words to enhance the probabilities of low-frequency words. This paper presents an extension to the word-pair method by involving multiple `predicting words' to produce better estimation for low-frequency words. We also employ this approach to deal with out-of-language words in the task of multi-lingual speech recognition.

Zhiyuan Tang, Dong Wang, Yiqiao Pan et al.

Pre-training is crucial for learning deep neural networks. Most of existing pre-training methods train simple models (e.g., restricted Boltzmann machines) and then stack them layer by layer to form the deep structure. This layer-wise pre-training has found strong theoretical foundation and broad empirical support. However, it is not easy to employ such method to pre-train models without a clear multi-layer structure,e.g., recurrent neural networks (RNNs). This paper presents a new pre-training approach based on knowledge transfer learning. In contrast to the layer-wise approach which trains model components incrementally, the new approach trains the entire model as a whole but with an easier objective function. This is achieved by utilizing soft targets produced by a prior trained model (teacher model). Compared to the conventional layer-wise methods, this new method does not care about the model structure, so can be used to pre-train very complex models. Experiments on a speech recognition task demonstrated that with this approach, complex RNNs can be well trained with a weaker deep neural network (DNN) model. Furthermore, the new method can be combined with conventional layer-wise pre-training to deliver additional gains.

Lantian Li, Dong Wang, Zhiyong Zhang et al.

Recent research shows that deep neural networks (DNNs) can be used to extract deep speaker vectors (d-vectors) that preserve speaker characteristics and can be used in speaker verification. This new method has been tested on text-dependent speaker verification tasks, and improvement was reported when combined with the conventional i-vector method. This paper extends the d-vector approach to semi text-independent speaker verification tasks, i.e., the text of the speech is in a limited set of short phrases. We explore various settings of the DNN structure used for d-vector extraction, and present a phone-dependent training which employs the posterior features obtained from an ASR system. The experimental results show that it is possible to apply d-vectors on semi text-independent speaker recognition, and the phone-dependent training improves system performance.

Zhiyuan Tang, Dong Wang, Zhiyong Zhang

Recurrent neural networks (RNNs), particularly long short-term memory (LSTM), have gained much attention in automatic speech recognition (ASR). Although some successful stories have been reported, training RNNs remains highly challenging, especially with limited training data. Recent research found that a well-trained model can be used as a teacher to train other child models, by using the predictions generated by the teacher model as supervision. This knowledge transfer learning has been employed to train simple neural nets with a complex one, so that the final performance can reach a level that is infeasible to obtain by regular training. In this paper, we employ the knowledge transfer learning approach to train RNNs (precisely LSTM) using a deep neural network (DNN) model as the teacher. This is different from most of the existing research on knowledge transfer learning, since the teacher (DNN) is assumed to be weaker than the child (RNN); however, our experiments on an ASR task showed that it works fairly well: without applying any tricks on the learning scheme, this approach can train RNNs successfully even with limited training data.