Learning from fully-unlabeled data is challenging in Multimedia Forensics problems, such as Person Re-Identification and Text Authorship Attribution. Recent self-supervised learning methods have shown to be effective when dealing with fully-unlabeled data in cases where the underlying classes have significant semantic differences, as intra-class distances are substantially lower than inter-class distances. However, this is not the case for forensic applications in which classes have similar semantics and the training and test sets have disjoint identities. General self-supervised learning methods might fail to learn discriminative features in this scenario, thus requiring more robust strategies. We propose a strategy to tackle Person Re-Identification and Text Authorship Attribution by enabling learning from unlabeled data even when samples from different classes are not prominently diverse. We propose a novel ensemble-based clustering strategy whereby clusters derived from different configurations are combined to generate a better grouping for the data samples in a fully-unsupervised way. This strategy allows clusters with different densities and higher variability to emerge, reducing intra-class discrepancies without requiring the burden of finding an optimal configuration per dataset. We also consider different Convolutional Neural Networks for feature extraction and subsequent distance computations between samples. We refine these distances by incorporating context and grouping them to capture complementary information. Our method is robust across both tasks, with different data modalities, and outperforms state-of-the-art methods with a fully-unsupervised solution without any labeling or human intervention.
相關內容
With the emergence of Cloud computing, Internet of Things-enabled Human-Computer Interfaces, Generative Artificial Intelligence, and high-accurate Machine and Deep-learning recognition and predictive models, along with the Post Covid-19 proliferation of social networking, and remote communications, the Metaverse gained a lot of popularity. Metaverse has the prospective to extend the physical world using virtual and augmented reality so the users can interact seamlessly with the real and virtual worlds using avatars and holograms. It has the potential to impact people in the way they interact on social media, collaborate in their work, perform marketing and business, teach, learn, and even access personalized healthcare. Several works in the literature examine Metaverse in terms of hardware wearable devices, and virtual reality gaming applications. However, the requirements of realizing the Metaverse in realtime and at a large-scale need yet to be examined for the technology to be usable. To address this limitation, this paper presents the temporal evolution of Metaverse definitions and captures its evolving requirements. Consequently, we provide insights into Metaverse requirements. In addition to enabling technologies, we lay out architectural elements for scalable, reliable, and efficient Metaverse systems, and a classification of existing Metaverse applications along with proposing required future research directions.
RGB-Thermal (RGB-T) pedestrian detection aims to locate the pedestrians in RGB-T image pairs to exploit the complementation between the two modalities for improving detection robustness in extreme conditions. Most existing algorithms assume that the RGB-T image pairs are well registered, while in the real world they are not aligned ideally due to parallax or different field-of-view of the cameras. The pedestrians in misaligned image pairs may locate at different positions in two images, which results in two challenges: 1) how to achieve inter-modality complementation using spatially misaligned RGB-T pedestrian patches, and 2) how to recognize the unpaired pedestrians at the boundary. To deal with these issues, we propose a new paradigm for unregistered RGB-T pedestrian detection, which predicts two separate pedestrian locations in the RGB and thermal images, respectively. Specifically, we propose a cross-modality proposal-guided feature mining (CPFM) mechanism to extract the two precise fusion features for representing the pedestrian in the two modalities, even if the RGB-T image pair is unaligned. It enables us to effectively exploit the complementation between the two modalities. With the CPFM mechanism, we build a two-stream dense detector; it predicts the two pedestrian locations in the two modalities based on the corresponding fusion feature mined by the CPFM mechanism. Besides, we design a data augmentation method, named Homography, to simulate the discrepancy in scales and views between images. We also investigate two non-maximum suppression (NMS) methods for post-processing. Favorable experimental results demonstrate the effectiveness and robustness of our method in dealing with unregistered pedestrians with different shifts.
Image Quality Assessment (IQA) constitutes a fundamental task within the field of computer vision, yet it remains an unresolved challenge, owing to the intricate distortion conditions, diverse image contents, and limited availability of data. Recently, the community has witnessed the emergence of numerous large-scale pretrained foundation models, which greatly benefit from dramatically increased data and parameter capacities. However, it remains an open problem whether the scaling law in high-level tasks is also applicable to IQA task which is closely related to low-level clues. In this paper, we demonstrate that with proper injection of local distortion features, a larger pretrained and fixed foundation model performs better in IQA tasks. Specifically, for the lack of local distortion structure and inductive bias of vision transformer (ViT), alongside the large-scale pretrained ViT, we use another pretrained convolution neural network (CNN), which is well known for capturing the local structure, to extract multi-scale image features. Further, we propose a local distortion extractor to obtain local distortion features from the pretrained CNN and a local distortion injector to inject the local distortion features into ViT. By only training the extractor and injector, our method can benefit from the rich knowledge in the powerful foundation models and achieve state-of-the-art performance on popular IQA datasets, indicating that IQA is not only a low-level problem but also benefits from stronger high-level features drawn from large-scale pretrained models.
Multimodal Named Entity Recognition (MNER) and Multimodal Relation Extraction (MRE) necessitate the fundamental reasoning capacity for intricate linguistic and multimodal comprehension. In this study, we explore distilling the reasoning ability of large language models (LLMs) into a more compact student model by generating a \textit{chain of thought} (CoT) -- a sequence of intermediate reasoning steps. Specifically, we commence by exemplifying the elicitation of such reasoning ability from LLMs through CoT prompts covering multi-grain (noun, sentence, multimodality) and data-augmentation (style, entity, image) dimensions. Subsequently, we present a novel conditional prompt distillation method to assimilate the commonsense reasoning ability from LLMs, thereby enhancing the utility of the student model in addressing text-only inputs without the requisite addition of image and CoT knowledge. Extensive experiments reveal that our approach attains state-of-the-art accuracy and manifests a plethora of advantages concerning interpretability, data efficiency, and cross-domain generalization on MNER and MRE datasets.
Cloth-changing person Re-IDentification (Re-ID) is a particularly challenging task, suffering from two limitations of inferior identity-relevant features and limited training samples. Existing methods mainly leverage auxiliary information to facilitate discriminative feature learning, including soft-biometrics features of shapes and gaits, and additional labels of clothing. However, these information may be unavailable in real-world applications. In this paper, we propose a novel FIne-grained Representation and Recomposition (FIRe$^{2}$) framework to tackle both limitations without any auxiliary information. Specifically, we first design a Fine-grained Feature Mining (FFM) module to separately cluster images of each person. Images with similar so-called fine-grained attributes (e.g., clothes and viewpoints) are encouraged to cluster together. An attribute-aware classification loss is introduced to perform fine-grained learning based on cluster labels, which are not shared among different people, promoting the model to learn identity-relevant features. Furthermore, by taking full advantage of the clustered fine-grained attributes, we present a Fine-grained Attribute Recomposition (FAR) module to recompose image features with different attributes in the latent space. It can significantly enhance representations for robust feature learning. Extensive experiments demonstrate that FIRe$^{2}$ can achieve state-of-the-art performance on five widely-used cloth-changing person Re-ID benchmarks.
Electronic Bill (E-Bill) is a rucial negotiable instrument in the form of data messages, relying on the Electronic Bill System (EB System). Blockchain technology offers inherent data sharing capabilities, so it is increasingly being adopted by small and medium-sized enterprises (SMEs) in the supply chain to build EB systems. However, the blockchain-based E-Bill still face significant challenges: the E-Bill is difficult to split, like non-fungible tokens (NFTs), and sensitive information such as amounts always be exposed on the blockchain. Therefore, to address these issues, we propose a novel data structure called Reverse-HashTree for Re-storing transactions in blockchain. In addition, we employ a variant of the Paillier public-key cryptosystem to ensure transaction validity without decryption, thus preserving privacy. Building upon these innovations, we designed BillChain, an EB system that enhances supply chain finance by providing privacy-preserving and splitting-enabled E-Bills on the blockchain. This work offers a comprehensive and innovative solution to the challenges faced by E-Bills applied in blockchain in the context of supply chain finance.
Synthesizing large logic programs through symbolic Inductive Logic Programming (ILP) typically requires intermediate definitions. However, cluttering the hypothesis space with intensional predicates typically degrades performance. In contrast, gradient descent provides an efficient way to find solutions within such high-dimensional spaces. Neuro-symbolic ILP approaches have not fully exploited this so far. We propose extending the {\delta}ILP approach to inductive synthesis with large-scale predicate invention, thus allowing us to exploit the efficacy of high-dimensional gradient descent. We show that large-scale predicate invention benefits differentiable inductive synthesis through gradient descent and allows one to learn solutions for tasks beyond the capabilities of existing neuro-symbolic ILP systems. Furthermore, we achieve these results without specifying the precise structure of the solution within the language bias.
Interest in the integration of Terrestrial Networks (TN) and Non-Terrestrial Networks (NTN); primarily satellites; has been rekindled due to the potential of NTN to provide ubiquitous coverage. Especially with the peculiar and flexible physical layer properties of 5G-NR, now direct access to 5G services through satellites could become possible. However, the large Round-Trip Delays (RTD) in NTNs require a re-evaluation of the design of RLC and PDCP layers timers ( and associated buffers), in particular for the regenerative payload satellites which have limited computational resources, and hence need to be optimally utilized. Our aim in this work is to initiate a new line of research for emerging NTNs with limited resources from a higher-layer perspective. To this end, we propose a novel and efficient method for optimally designing the RLC and PDCP layers' buffers and timers without the need for intensive computations. This approach is relevant for low-cost satellites, which have limited computational and energy resources. The simulation results show that the proposed methods can significantly improve the performance in terms of resource utilization and delays.
We propose a novel constrained Bayesian Optimization (BO) algorithm optimizing the design process of Laterally-Diffused Metal-Oxide-Semiconductor (LDMOS) transistors while realizing a target Breakdown Voltage (BV). We convert the constrained BO problem into a conventional BO problem using a Lagrange multiplier. Instead of directly optimizing the traditional Figure-of-Merit (FOM), we set the Lagrangian as the objective function of BO. This adaptive objective function with a changeable Lagrange multiplier can address constrained BO problems which have constraints that require costly evaluations, without the need for additional surrogate models to approximate constraints. Our algorithm enables a device designer to set the target BV in the design space, and obtain a device that satisfies the optimized FOM and the target BV constraint automatically. Utilizing this algorithm, we have also explored the physical limits of the FOM for our devices in 30 - 50 V range within the defined design space.
Reasoning with knowledge expressed in natural language and Knowledge Bases (KBs) is a major challenge for Artificial Intelligence, with applications in machine reading, dialogue, and question answering. General neural architectures that jointly learn representations and transformations of text are very data-inefficient, and it is hard to analyse their reasoning process. These issues are addressed by end-to-end differentiable reasoning systems such as Neural Theorem Provers (NTPs), although they can only be used with small-scale symbolic KBs. In this paper we first propose Greedy NTPs (GNTPs), an extension to NTPs addressing their complexity and scalability limitations, thus making them applicable to real-world datasets. This result is achieved by dynamically constructing the computation graph of NTPs and including only the most promising proof paths during inference, thus obtaining orders of magnitude more efficient models. Then, we propose a novel approach for jointly reasoning over KBs and textual mentions, by embedding logic facts and natural language sentences in a shared embedding space. We show that GNTPs perform on par with NTPs at a fraction of their cost while achieving competitive link prediction results on large datasets, providing explanations for predictions, and inducing interpretable models. Source code, datasets, and supplementary material are available online at //github.com/uclnlp/gntp.
小貼士
登錄享
相關主題
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191