Zhao et al.Original Research
Graph Distribution-valued Signals: A Wasserstein Space Perspective
Novel framework models signals as graph distribution-valued signals, suggesting improved handling of uncertainty in predictions.
41 pubs · 148 cites · h-index 6
Zhao et al.Original Research
Novel framework models signals as graph distribution-valued signals, suggesting improved handling of uncertainty in predictions.
Zhao et al.Original Research
This framework enhances graph contrastive learning without negative samples, suggesting better performance across datasets.
Zhao et al.Original Research
We introduce a novel uncertainty principle for generalized graph signals that extends classical time-frequency and graph uncertainty principles into a unified framework. By defining joint vertex-time and spectral-frequency spreads, we quantify signal localization across these domains, revealing a trade-off between them. This framework allows us to identify a class of signals with maximal energy concentration in both domains, forming the fundamental atoms for a new joint vertex-time dictionary. This dictionary enhances signal reconstruction under practical constraints, such as intermittent data, commonly encountered in sensor and social networks. Numerical experiments on real-world datasets demonstrate the effectiveness of the proposed approach, showing improved reconstruction accuracy and noise robustness compared to existing methods.
IEEE Transactions on Signal ProcessingJian et al.Original Research
The detection of interesting or anomalous signal behavior using sensor networks plays a key role in many applications. In this work, we model the sensor network as a graph, with each vertex representing a sensor and a signal over time associated with each vertex. The objective is to identify the true state of the signal at each point in the joint spatio-temporal domain. We propose a step-up empirical Bayes multiple hypothesis testing approach to make decisions based on local summary statistics. To this end, we establish consistent estimates of the prior probability of the null hypothesis as well as the probability models under the alternative, which are obtained using a bandlimited generalized graph signal model. Asymptotic control of the false discovery rate is proven. Numerical experiments validate the effectiveness of our approach compared to existing methods.
IEEE Transactions on Geoscience and Remote SensingLuo et al.Original Research
Ground-penetrating radar (GPR) is a vital tool in the domain of nondestructive testing; however, its capability to accurately discern subsurface targets faces challenges from substantial background clutter. Current methods aimed at clutter suppression often leave residual clutter or distort the hyperbolic tails of target-scattered signals, particularly in heterogeneous soil conditions. This study endeavors to tackle the complexities of clutter suppression in practical scenarios. To this end, we introduce a novel framework for GPR clutter suppression using stationary graph signal (SGS) processing techniques. In our proposed approach, GPR B-scan images are treated as graph signals and transformed into the graph frequency domain via graph Fourier transform (GFT). This framework incorporates B-scan images containing both targets and clutter alongside clutter-only B-scan images gathered within the same testing environment. B-scan images featuring both targets and clutter serve as reference data samples for constructing a graph shift operator (GSO), with clutter and targets’ scattering signals interpreted as SGS. Following the establishment of weak SGSs with respect to the GSO, a variant of the graph-based Wiener filter tailored for GPR applications is applied to effectuate clutter suppression. Through our proposed SGS processing-based filtering method, clutter can be effectively suppressed, thereby facilitating the restoration of target scattering signals. Extensive experiments conducted on both numerical simulation data and field test data underscore the efficacy of the proposed approach, which can be further applied to the general nondestructive testing realm.
Zhang et al.Original Research
Topological signal processing (TSP) utilizes simplicial complexes to model structures with higher order than vertices and edges. In this paper, we study the transferability of TSP via a generalized higher-order version of graphon, known as complexon. We recall the notion of a complexon as the limit of a simplicial complex sequence [1]. Inspired by the graphon shift operator and message-passing neural network, we construct a marginal complexon and complexon shift operator (CSO) according to components of all possible dimensions from the complexon. We investigate the CSO's eigenvalues and eigenvectors and relate them to a new family of weighted adjacency matrices. We prove that when a simplicial complex signal sequence converges to a complexon signal, the eigenvalues, eigenspaces, and Fourier transform of the corresponding CSOs converge to that of the limit complexon signal. This conclusion is further verified by two numerical experiments. These results hint at learning transferability on large simplicial complexes or simplicial complex sequences, which generalize the graphon signal processing framework.
Wang et al.Original Research
The utilization of multi-modal sensor data in visual place recognition (VPR) has demonstrated enhanced performance compared to single-modal counterparts. Nonetheless, integrating additional sensors comes with elevated costs and may not be feasible for systems that demand lightweight operation, thereby impacting the practical deployment of VPR. To address this issue, we resort to knowledge distillation, which empowers single-modal students to learn from cross-modal teachers without introducing additional sensors during inference. Despite the notable advancements achieved by current distillation approaches, the exploration of feature relationships remains an under-explored area. In order to tackle the challenge of cross-modal distillation in VPR, we present DistilVPR, a novel distillation pipeline for VPR. We propose leveraging feature relationships from multiple agents, including self-agents and cross-agents for teacher and student neural networks. Furthermore, we integrate various manifolds, characterized by different space curvatures for exploring feature relationships. This approach enhances the diversity of feature relationships, including Euclidean, spherical, and hyperbolic relationship modules, thereby enhancing the overall representational capacity. The experiments demonstrate that our proposed pipeline achieves state-of-the-art performance compared to other distillation baselines. We also conduct necessary ablation studies to show design effectiveness. The code is released at: https://github.com/sijieaaa/DistilVPR
She et al.Original Research
Point cloud registration is a crucial technique in 3D computer vision with a wide range of applications. However, this task can be challenging, particularly in large fields of view with dynamic objects, environmental noise, or other perturbations. To address this challenge, we propose a model called PosDiffNet. Our approach performs hierarchical registration based on window-level, patch-level, and point-level correspondence. We leverage a graph neural partial differential equation (PDE) based on Beltrami flow to obtain high-dimensional features and position embeddings for point clouds. We incorporate position embeddings into a Transformer module based on a neural ordinary differential equation (ODE) to efficiently represent patches within points. We employ the multi-level correspondence derived from the high feature similarity scores to facilitate alignment between point clouds. Subsequently, we use registration methods such as SVD-based algorithms to predict the transformation using corresponding point pairs. We evaluate PosDiffNet on several 3D point cloud datasets, verifying that it achieves state-of-the-art (SOTA) performance for point cloud registration in large fields of view with perturbations. The implementation code of experiments is available at https://github.com/AI-IT-AVs/PosDiffNet.
IEEE Transactions on Signal ProcessingJi et al.Original Research
Graphons are limit objects of sequences of graphs and are used to analyze the behavior of large graphs. Recently, graphon signal processing has been developed to study signal processing on large graphs. A major limitation of this approach is that any sparse sequence of graphs inevitably converges to the zero graphon, rendering the resulting signal processing theory trivial and inadequate for sparse graph sequences. To overcome this limitation, we propose a new signal processing framework that leverages the concept of generalized graphons and introduces the stretched cut distance as a measure to compare these graphons. Our framework focuses on the sampling of graph sequences from generalized graphons and explores the convergence properties of associated operators, spectra, and signals. Our signal processing framework provides a comprehensive approach to analyzing and processing signals on graph sequences, even if they are sparse. Finally, we discuss the practical implications of our theory for real-world large networks through numerical experiments.
IEEE Transactions on Signal ProcessingJian et al.Original Research
In generalized graph signal processing (GGSP), the signal associated with each vertex in a graph is an element from a Hilbert space. In this paper, we study GGSP signal reconstruction as a kernel ridge regression (KRR) problem. By devising an appropriate kernel, we show that this problem has a solution that can be evaluated in a distributed way. We interpret the problem and solution using both deterministic and Bayesian perspectives and link them to existing graph signal processing and GGSP frameworks. We then provide an online implementation via random Fourier features. Under the Bayesian framework, we investigate the statistical performance under the asymptotic sampling scheme. Finally, we validate our theory and methods on real-world datasets.