Currently, low-resolution image recognition is confronted with a significant challenge in the field of intelligent traffic perception. Compared to high-resolution images, low-resolution images suffer from small size, low quality, and lack of detail, leading to a notable decrease in the accuracy of traditional neural network recognition algorithms. The key to low-resolution image recognition lies in effective feature extraction. Therefore, this paper delves into the fundamental dimensions of residual modules and their impact on feature extraction and computational efficiency. Based on experiments, we introduce a dual-branch residual network structure that leverages the basic architecture of residual networks and a common feature subspace algorithm. Additionally, it incorporates the utilization of intermediate-layer features to enhance the accuracy of low-resolution image recognition. Furthermore, we employ knowledge distillation to reduce network parameters and computational overhead. Experimental results validate the effectiveness of this algorithm for low-resolution image recognition in traffic environments.

This PhD thesis is focused on the central idea that single neurons in the brain should be regarded as temporally precise and highly complex spatio-temporal pattern recognizers. This is opposed to the prevalent view of biological neurons as simple and mainly spatial pattern recognizers by most neuroscientists today. In this thesis, I will attempt to demonstrate that this is an important distinction, predominantly because the above-mentioned computational properties of single neurons have far-reaching implications with respect to the various brain circuits that neurons compose, and on how information is encoded by neuronal activity in the brain. Namely, that these particular "low-level" details at the single neuron level have substantial system-wide ramifications. In the introduction we will highlight the main components that comprise a neural microcircuit that can perform useful computations and illustrate the inter-dependence of these components from a system perspective. In chapter 1 we discuss the great complexity of the spatio-temporal input-output relationship of cortical neurons that are the result of morphological structure and biophysical properties of the neuron. In chapter 2 we demonstrate that single neurons can generate temporally precise output patterns in response to specific spatio-temporal input patterns with a very simple biologically plausible learning rule. In chapter 3, we use the differentiable deep network analog of a realistic cortical neuron as a tool to approximate the gradient of the output of the neuron with respect to its input and use this capability in an attempt to teach the neuron to perform nonlinear XOR operation. In chapter 4 we expand chapter 3 to describe extension of our ideas to neuronal networks composed of many realistic biological spiking neurons that represent either small microcircuits or entire brain regions.

The proliferation and ubiquity of temporal data across many disciplines has sparked interest for similarity, classification and clustering methods specifically designed to handle time series data. A core issue when dealing with time series is determining their pairwise similarity, i.e., the degree to which a given time series resembles another. Traditional distance measures such as the Euclidean are not well-suited due to the time-dependent nature of the data. Elastic metrics such as dynamic time warping (DTW) offer a promising approach, but are limited by their computational complexity, non-differentiability and sensitivity to noise and outliers. This thesis proposes novel elastic alignment methods that use parametric \& diffeomorphic warping transformations as a means of overcoming the shortcomings of DTW-based metrics. The proposed method is differentiable \& invertible, well-suited for deep learning architectures, robust to noise and outliers, computationally efficient, and is expressive and flexible enough to capture complex patterns. Furthermore, a closed-form solution was developed for the gradient of these diffeomorphic transformations, which allows an efficient search in the parameter space, leading to better solutions at convergence. Leveraging the benefits of these closed-form diffeomorphic transformations, this thesis proposes a suite of advancements that include: (a) an enhanced temporal transformer network for time series alignment and averaging, (b) a deep-learning based time series classification model to simultaneously align and classify signals with high accuracy, (c) an incremental time series clustering algorithm that is warping-invariant, scalable and can operate under limited computational and time resources, and finally, (d) a normalizing flow model that enhances the flexibility of affine transformations in coupling and autoregressive layers.

Hand gesture recognition (HGR) has gained significant attention due to the increasing use of AI-powered human-computer interfaces that can interpret the deep spatiotemporal dynamics of biosignals from the peripheral nervous system, such as surface electromyography (sEMG). These interfaces have a range of applications, including the control of extended reality, agile prosthetics, and exoskeletons. However, the natural variability of sEMG among individuals has led researchers to focus on subject-specific solutions. Deep learning methods, which often have complex structures, are particularly data-hungry and can be time-consuming to train, making them less practical for subject-specific applications. In this paper, we propose and develop a generalizable, sequential decoder of transient high-density sEMG (HD-sEMG) that achieves 73% average accuracy on 65 gestures for partially-observed subjects through subject-embedded transfer learning, leveraging pre-knowledge of HGR acquired during pre-training. The use of transient HD-sEMG before gesture stabilization allows us to predict gestures with the ultimate goal of counterbalancing system control delays. The results show that the proposed generalized models significantly outperform subject-specific approaches, especially when the training data is limited, and there is a significant number of gesture classes. By building on pre-knowledge and incorporating a multiplicative subject-embedded structure, our method comparatively achieves more than 13% average accuracy across partially observed subjects with minimal data availability. This work highlights the potential of HD-sEMG and demonstrates the benefits of modeling common patterns across users to reduce the need for large amounts of data for new users, enhancing practicality.

Current adversarial attacks against speaker recognition systems (SRSs) require either white-box access or heavy black-box queries to the target SRS, thus still falling behind practical attacks against proprietary commercial APIs and voice-controlled devices. To fill this gap, we propose QFA2SR, an effective and imperceptible query-free black-box attack, by leveraging the transferability of adversarial voices. To improve transferability, we present three novel methods, tailored loss functions, SRS ensemble, and time-freq corrosion. The first one tailors loss functions to different attack scenarios. The latter two augment surrogate SRSs in two different ways. SRS ensemble combines diverse surrogate SRSs with new strategies, amenable to the unique scoring characteristics of SRSs. Time-freq corrosion augments surrogate SRSs by incorporating well-designed time-/frequency-domain modification functions, which simulate and approximate the decision boundary of the target SRS and distortions introduced during over-the-air attacks. QFA2SR boosts the targeted transferability by 20.9%-70.7% on four popular commercial APIs (Microsoft Azure, iFlytek, Jingdong, and TalentedSoft), significantly outperforming existing attacks in query-free setting, with negligible effect on the imperceptibility. QFA2SR is also highly effective when launched over the air against three wide-spread voice assistants (Google Assistant, Apple Siri, and TMall Genie) with 60%, 46%, and 70% targeted transferability, respectively.

Nowadays, increasingly more data are available as knowledge graphs (KGs). While this data model supports advanced reasoning and querying, they remain difficult to mine due to their size and complexity. Graph mining approaches can be used to extract patterns from KGs. However this presents two main issues. First, graph mining approaches tend to extract too many patterns for a human analyst to interpret (pattern explosion). Second, real-life KGs tend to differ from the graphs usually treated in graph mining: they are multigraphs, their vertex degrees tend to follow a power-law, and the way in which they model knowledge can produce spurious patterns. Recently, a graph mining approach named GraphMDL+ has been proposed to tackle the problem of pattern explosion, using the Minimum Description Length (MDL) principle. However, GraphMDL+, like other graph mining approaches, is not suited for KGs without adaptations. In this paper we propose KG-MDL, a graph pattern mining approach based on the MDL principle that, given a KG, generates a human-sized and descriptive set of graph patterns, and so in a parameter-less and anytime way. We report on experiments on medium-sized KGs showing that our approach generates sets of patterns that are both small enough to be interpreted by humans and descriptive of the KG. We show that the extracted patterns highlight relevant characteristics of the data: both of the schema used to create the data, and of the concrete facts it contains. We also discuss the issues related to mining graph patterns on knowledge graphs, as opposed to other types of graph data.

Over the last decades, ample achievements have been made on Structure from motion (SfM). However, the vast majority of them basically work in an offline manner, i.e., images are firstly captured and then fed together into a SfM pipeline for obtaining poses and sparse point cloud. In this work, on the contrary, we present an on-the-fly SfM: running online SfM while image capturing, the newly taken On-the-Fly image is online estimated with the corresponding pose and points, i.e., what you capture is what you get. Specifically, our approach firstly employs a vocabulary tree that is unsupervised trained using learning-based global features for fast image retrieval of newly fly-in image. Then, a robust feature matching mechanism with least squares (LSM) is presented to improve image registration performance. Finally, via investigating the influence of newly fly-in image's connected neighboring images, an efficient hierarchical weighted local bundle adjustment (BA) is used for optimization. Extensive experimental results demonstrate that on-the-fly SfM can meet the goal of robustly registering the images while capturing in an online way.

This paper explores semi-supervised training for sequence tasks, such as Optical Character Recognition or Automatic Speech Recognition. We propose a novel loss function $\unicode{x2013}$ SoftCTC $\unicode{x2013}$ which is an extension of CTC allowing to consider multiple transcription variants at the same time. This allows to omit the confidence based filtering step which is otherwise a crucial component of pseudo-labeling approaches to semi-supervised learning. We demonstrate the effectiveness of our method on a challenging handwriting recognition task and conclude that SoftCTC matches the performance of a finely-tuned filtering based pipeline. We also evaluated SoftCTC in terms of computational efficiency, concluding that it is significantly more efficient than a na\"ive CTC-based approach for training on multiple transcription variants, and we make our GPU implementation public.

Medical image registration is a critical task that estimates the spatial correspondence between pairs of images. However, current traditional and deep-learning-based methods rely on similarity measures to generate a deforming field, which often results in disproportionate volume changes in dissimilar regions, especially in tumor regions. These changes can significantly alter the tumor size and underlying anatomy, which limits the practical use of image registration in clinical diagnosis. To address this issue, we have formulated image registration with tumors as a constraint problem that preserves tumor volumes while maximizing image similarity in other normal regions. Our proposed strategy involves a two-stage process. In the first stage, we use similarity-based registration to identify potential tumor regions by their volume change, generating a soft tumor mask accordingly. In the second stage, we propose a volume-preserving registration with a novel adaptive volume-preserving loss that penalizes the change in size adaptively based on the masks calculated from the previous stage. Our approach balances image similarity and volume preservation in different regions, i.e., normal and tumor regions, by using soft tumor masks to adjust the imposition of volume-preserving loss on each one. This ensures that the tumor volume is preserved during the registration process. We have evaluated our strategy on various datasets and network architectures, demonstrating that our method successfully preserves the tumor volume while achieving comparable registration results with state-of-the-art methods. Our codes is available at: \url{https://dddraxxx.github.io/Volume-Preserving-Registration/}.

Gait recognition (GR) is a growing biometric modality used for person identification from a distance through visual cameras. GR provides a secure and reliable alternative to fingerprint and face recognition, as it is harder to distinguish between false and authentic signals. Furthermore, its resistance to spoofing makes GR suitable for all types of environments. With the rise of deep learning, steadily improving strides have been made in GR technology with promising results in various contexts. As video surveillance becomes more prevalent, new obstacles arise, such as ensuring uniform performance evaluation across different protocols, reliable recognition despite shifting lighting conditions, fluctuations in gait patterns, and protecting privacy.This survey aims to give an overview of GR and analyze the environmental elements and complications that could affect it in comparison to other biometric recognition systems. The primary goal is to examine the existing deep learning (DL) techniques employed for human GR that may generate new research opportunities.