Search

Energy based models (EBMs) are appealing for their generality and simplicity in data likelihood modeling, but have conventionally been difficult to train due to the unstable and time-consuming implicit MCMC sampling during contrastive divergence training. In this paper, we present a novel energy-based generative framework, Variational Potential Flow (VAPO), that entirely dispenses with implicit MCMC sampling and does not rely on complementary latent models or cooperative training. The VAPO framework aims to learn a potential energy function whose gradient (flow) guides the prior samples, so that their density evolution closely follows an approximate data likelihood homotopy. An energy loss function is then formulated to minimize the Kullback-Leibler divergence between density evolution of the flow-driven prior and the data likelihood homotopy. Images can be generated after training the potential energy, by initializing the samples from Gaussian prior and solving the ODE governing the potential flow on a fixed time interval using generic ODE solvers. Experiment results show that the proposed VAPO framework is capable of generating realistic images on various image datasets. In particular, our proposed framework achieves competitive FID scores for unconditional image generation on the CIFAR-10 and CelebA datasets.

Existing brain tumor segmentation methods usually utilize multiple Magnetic Resonance Imaging (MRI) modalities in brain tumor images for segmentation, which can achieve better segmentation performance. However, in clinical applications, some modalities are missing due to resource constraints, leading to severe degradation in the performance of methods applying complete modality segmentation. In this paper, we propose a Multimodal feature distillation with Convolutional Neural Network (CNN)-Transformer hybrid network (MCTSeg) for accurate brain tumor segmentation with missing modalities. We first design a Multimodal Feature Distillation (MFD) module to distill feature-level multimodal knowledge into different unimodality to extract complete modality information. We further develop a Unimodal Feature Enhancement (UFE) module to model the relationship between global and local information semantically. Finally, we build a Cross-Modal Fusion (CMF) module to explicitly align the global correlations among different modalities even when some modalities are missing. Complementary features within and across different modalities are refined via the CNN-Transformer hybrid architectures in both the UFE and CMF modules, where local and global dependencies are both captured. Our ablation study demonstrates the importance of the proposed modules with CNN-Transformer networks and the convolutional blocks in Transformer for improving the performance of brain tumor segmentation with missing modalities. Extensive experiments on the BraTS2018 and BraTS2020 datasets show that the proposed MCTSeg framework outperforms the state-of-the-art methods in missing modalities cases. Our code is available at: https://github.com/mkang315/MCTSeg.

Human action or activity recognition in videos is a fundamental task in computer vision with applications in surveillance and monitoring, self-driving cars, sports analytics, human-robot interaction and many more. Traditional supervised methods require large annotated datasets for training, which are expensive and time-consuming to acquire. This work proposes a novel approach using Cross-Architecture Pseudo-Labeling with contrastive learning for semi-supervised action recognition. Our framework leverages both labeled and unlabelled data to robustly learn action representations in videos, combining pseudo-labeling with contrastive learning for effective learning from both types of samples. We introduce a novel cross-architecture approach where 3D Convolutional Neural Networks (3D CNNs) and video transformers (VIT) are utilised to capture different aspects of action representations; hence we call it ActNetFormer. The 3D CNNs excel at capturing spatial features and local dependencies in the temporal domain, while VIT excels at capturing long-range dependencies across frames. By integrating these complementary architectures within the ActNetFormer framework, our approach can effectively capture both local and global contextual information of an action. This comprehensive representation learning enables the model to achieve better performance in semi-supervised action recognition tasks by leveraging the strengths of each of these architectures. Experimental results on standard action recognition datasets demonstrate that our approach performs better than the existing methods, achieving state-of-the-art performance with only a fraction of labeled data. The official website of this work is available at: https://github.com/rana2149/ActNetFormer., Comment: Submitted for peer review

The low-rank plus sparse (L+S) decomposition model has enabled better reconstruction of dynamic magnetic resonance imaging (dMRI) with separation into background (L) and dynamic (S) component. However, use of low-rank prior alone may not fully explain the slow variations or smoothness of the background part at the local scale. In this paper, we propose a smoothness-regularized L+S (SR-L+S) model for dMRI reconstruction from highly undersampled k-t-space data. We exploit joint low-rank and smooth priors on the background component of dMRI to better capture both its global and local temporal correlated structures. Extending the L+S formulation, the low-rank property is encoded by the nuclear norm, while the smoothness by a general \ell_{p}-norm penalty on the local differences of the columns of L. The additional smoothness regularizer can promote piecewise local consistency between neighboring frames. By smoothing out the noise and dynamic activities, it allows accurate recovery of the background part, and subsequently more robust dMRI reconstruction. Extensive experiments on multi-coil cardiac and synthetic data shows that the SR-L+S model outp, Comment: 9 pages

We propose a novel Attentional Scale Sequence Fusion based You Only Look Once (YOLO) framework (ASF-YOLO) which combines spatial and scale features for accurate and fast cell instance segmentation. Built on the YOLO segmentation framework, we employ the Scale Sequence Feature Fusion (SSFF) module to enhance the multi-scale information extraction capability of the network, and the Triple Feature Encoder (TFE) module to fuse feature maps of different scales to increase detailed information. We further introduce a Channel and Position Attention Mechanism (CPAM) to integrate both the SSFF and TPE modules, which focus on informative channels and spatial position-related small objects for improved detection and segmentation performance. Experimental validations on two cell datasets show remarkable segmentation accuracy and speed of the proposed ASF-YOLO model. It achieves a box mAP of 0.91, mask mAP of 0.887, and an inference speed of 47.3 FPS on the 2018 Data Science Bowl dataset, outperforming the state-of-the-art methods. The source code is available at https://github.com/mkang315/ASF-YOLO.

You Only Look Once (YOLO)-based object detectors have shown remarkable accuracy for automated brain tumor detection. In this paper, we develop a novel BGF-YOLO architecture by incorporating Bi-level Routing Attention (BRA), Generalized feature pyramid networks (GFPN), and Fourth detecting head into YOLOv8. BGF-YOLO contains an attention mechanism to focus more on important features, and feature pyramid networks to enrich feature representation by merging high-level semantic features with spatial details. Furthermore, we investigate the effect of different attention mechanisms and feature fusions, detection head architectures on brain tumor detection accuracy. Experimental results show that BGF-YOLO gives a 4.7% absolute increase of mAP$_{50}$ compared to YOLOv8x, and achieves state-of-the-art on the brain tumor detection dataset Br35H. The code is available at https://github.com/mkang315/BGF-YOLO.

With an excellent balance between speed and accuracy, cutting-edge YOLO frameworks have become one of the most efficient algorithms for object detection. However, the performance of using YOLO networks is scarcely investigated in brain tumor detection. We propose a novel YOLO architecture with Reparameterized Convolution based on channel Shuffle (RCS-YOLO). We present RCS and a One-Shot Aggregation of RCS (RCS-OSA), which link feature cascade and computation efficiency to extract richer information and reduce time consumption. Experimental results on the brain tumor dataset Br35H show that the proposed model surpasses YOLOv6, YOLOv7, and YOLOv8 in speed and accuracy. Notably, compared with YOLOv7, the precision of RCS-YOLO improves by 1%, and the inference speed by 60% at 114.8 images detected per second (FPS). Our proposed RCS-YOLO achieves state-of-the-art performance on the brain tumor detection task. The code is available at https://github.com/mkang315/RCS-YOLO.

Recent applications of pattern recognition techniques on brain connectome classification using functional connectivity (FC) are shifting towards acknowledging the non-Euclidean topology and causal dynamics of brain connectivity across time. In this paper, a deep spatiotemporal variational Bayes (DSVB) framework is proposed to learn time-varying topological structures in dynamic FC networks for identifying autism spectrum disorder (ASD) in human participants. The framework incorporates a spatial-aware recurrent neural network with an attention-based message passing scheme to capture rich spatiotemporal patterns across dynamic FC networks. To overcome model overfitting on limited training datasets, an adversarial training strategy is introduced to learn graph embedding models that generalize well to unseen brain networks. Evaluation on the ABIDE resting-state functional magnetic resonance imaging dataset shows that our proposed framework substantially outperforms state-of-the-art methods in identifying patients with ASD. Dynamic FC analyses with DSVB-learned embeddings reveal apparent group differences between ASD and healthy controls in brain network connectivity patterns and switching dynamics of brain states.

Common measures of brain functional connectivity (FC) including covariance and correlation matrices are semi-positive definite (SPD) matrices residing on a cone-shape Riemannian manifold. Despite its remarkable success for Euclidean-valued data generation, use of standard generative adversarial networks (GANs) to generate manifold-valued FC data neglects its inherent SPD structure and hence the inter-relatedness of edges in real FC. We propose a novel graph-regularized manifold-aware conditional Wasserstein GAN (GR-SPD-GAN) for FC data generation on the SPD manifold that can preserve the global FC structure. Specifically, we optimize a generalized Wasserstein distance between the real and generated SPD data under an adversarial training, conditioned on the class labels. The resulting generator can synthesize new SPD-valued FC matrices associated with different classes of brain networks, e.g., brain disorder or healthy control. Furthermore, we introduce additional population graph-based regularization terms on both the SPD manifold and its tangent space to encourage the generator to respect the inter-subject similarity of FC patterns in the real data. This also helps in avoiding mode collapse and produces more stable GAN training. Evaluated on resting-state functional magnetic resonance imaging (fMRI) data of major depressive disorder (MDD), qualitative and quantitative results show that the proposed GR-SPD-GAN clearly outperforms several state-of-the-art GANs in generating more realistic fMRI-based FC samples. When applied to FC data augmentation for MDD identification, classification models trained on augmented data generated by our approach achieved the largest margin of improvement in classification accuracy among the competing GANs over baselines without data augmentation., Comment: 10 pages, 4 figures

GAN-generated deepfakes as a genre of digital images are gaining ground as both catalysts of artistic expression and malicious forms of deception, therefore demanding systems to enforce and accredit their ethical use. Existing techniques for the source attribution of synthetic images identify subtle intrinsic fingerprints using multiclass classification neural nets limited in functionality and scalability. Hence, we redefine the deepfake detection and source attribution problems as a series of related binary classification tasks. We leverage transfer learning to rapidly adapt forgery detection networks for multiple independent attribution problems, by proposing a semi-decentralized modular design to solve them simultaneously and efficiently. Class activation mapping is also demonstrated as an effective means of feature localization for model interpretation. Our models are determined via experimentation to be competitive with current benchmarks, and capable of decent performance on human portraits in ideal conditions. Decentralized fingerprint-based attribution is found to retain validity in the presence of novel sources, but is more susceptible to type II errors that intensify with image perturbations and attributive uncertainty. We describe both our conceptual framework and model prototypes for further enhancement when investigating the technical limits of reactive deepfake attribution., Comment: 21 pages, 8 figures. Code: https://github.com/quarxilon/Generator_Attribution

Brain functional connectivity (FC) reveals biomarkers for identification of various neuropsychiatric disorders. Recent application of deep neural networks (DNNs) to connectome-based classification mostly relies on traditional convolutional neural networks using input connectivity matrices on a regular Euclidean grid. We propose a graph deep learning framework to incorporate the non-Euclidean information about graph structure for classifying functional magnetic resonance imaging (fMRI)-derived brain networks in major depressive disorder (MDD). We design a novel graph autoencoder (GAE) architecture based on the graph convolutional networks (GCNs) to embed the topological structure and node content of large-sized fMRI networks into low-dimensional latent representations. In network construction, we employ the Ledoit-Wolf (LDW) shrinkage method to estimate the high-dimensional FC metrics efficiently from fMRI data. We consider both supervised and unsupervised approaches for the graph embedding learning. The learned embeddings are then used as feature inputs for a deep fully-connected neural network (FCNN) to discriminate MDD from healthy controls. Evaluated on two resting-state fMRI (rs-fMRI) MDD datasets, results show that the proposed GAE-FCNN model significantly outperforms several state-of-the-art methods for brain connectome classification, achieving the best accuracy using the LDW-FC edges as node features. The graph embeddings of fMRI FC networks learned by the GAE also reveal apparent group differences between MDD and HC. Our new framework demonstrates feasibility of learning graph embeddings on brain networks to provide discriminative information for diagnosis of brain disorders.

Facial micro-expression (ME) recognition has posed a huge challenge to researchers for its subtlety in motion and limited databases. Recently, handcrafted techniques have achieved superior performance in micro-expression recognition but at the cost of domain specificity and cumbersome parametric tunings. In this paper, we propose an Enriched Long-term Recurrent Convolutional Network (ELRCN) that first encodes each micro-expression frame into a feature vector through CNN module(s), then predicts the micro-expression by passing the feature vector through a Long Short-term Memory (LSTM) module. The framework contains two different network variants: (1) Channel-wise stacking of input data for spatial enrichment, (2) Feature-wise stacking of features for temporal enrichment. We demonstrate that the proposed approach is able to achieve reasonably good performance, without data augmentation. In addition, we also present ablation studies conducted on the framework and visualizations of what CNN "sees" when predicting the micro-expression classes., Comment: Published in Micro-Expression Grand Challenge 2018, Workshop of 13th IEEE Facial & Gesture 2018

Despite recent interest and advances in facial micro-expression research, there is still plenty room for improvement in terms of micro-expression recognition. Conventional feature extraction approaches for micro-expression video consider either the whole video sequence or a part of it, for representation. However, with the high-speed video capture of micro-expressions (100-200 fps), are all frames necessary to provide a sufficiently meaningful representation? Is the luxury of data a bane to accurate recognition? A novel proposition is presented in this paper, whereby we utilize only two images per video: the apex frame and the onset frame. The apex frame of a video contains the highest intensity of expression changes among all frames, while the onset is the perfect choice of a reference frame with neutral expression. A new feature extractor, Bi-Weighted Oriented Optical Flow (Bi-WOOF) is proposed to encode essential expressiveness of the apex frame. We evaluated the proposed method on five micro-expression databases: CAS(ME)$^2$, CASME II, SMIC-HS, SMIC-NIR and SMIC-VIS. Our experiments lend credence to our hypothesis, with our proposed technique achieving a state-of-the-art F1-score recognition performance of 61% and 62% in the high frame rate CASME II and SMIC-HS databases respectively., Comment: 14 pages double-column, author affiliations updated, acknowledgment of grant support added

Higher order differentiation was introduced in a cryptographic context by Lai. Several attacks can be viewed in the context of higher order differentiations, amongst them the cube attack and the AIDA attack. All of the above have been developed for the binary case. We examine differentiation in larger fields, starting with the field $GF(p)$ of integers modulo a prime $p$. We prove a number of results on differentiating polynomials over such fields and then apply these techniques to generalising the cube attack to $GF(p)$. The crucial difference is that now the degree in each variable can be higher than one, and our proposed attack will differentiate several times with respect to each variable (unlike the classical cube attack and its larger field version described by Dinur and Shamir, both of which differentiate at most once with respect to each variable). Finally we describe differentiation over finite fields $GF(p^m)$ with $p^m$ elements and prove that it can be reduced to differentiation over $GF(p)$, so a cube attack over $GF(p^m)$ would be equivalent to cube attacks over $GF(p)$., Comment: submitted to a journal

To date, most facial expression analysis have been based on posed image databases and is carried out without being able to protect the identity of the subjects whose expressions are being recognised. In this paper, we propose and implement a system for classifying facial expressions of images in the encrypted domain based on a Paillier cryptosystem implementation of Fisher Linear Discriminant Analysis and k-nearest neighbour (FLDA + kNN). We present results of experiments carried out on a recently developed natural visible and infrared facial expression (NVIE) database of spontaneous images. To the best of our knowledge, this is the first system that will allow the recog-nition of encrypted spontaneous facial expressions by a remote server on behalf of a client., Comment: 4 pages. 9th IMA International Conference on Mathematics in Signal Processing, Birmingham, UK, Dec. 2012