Few-shot image classification is a critical issue in the field of computer vision, facing challenges related to data scarcity and model generalization. Transformer models, representing self-attention mechanisms, have ...
详细信息
ISBN:
(数字)9798350349115
ISBN:
(纸本)9798350349122
Few-shot image classification is a critical issue in the field of computer vision, facing challenges related to data scarcity and model generalization. Transformer models, representing self-attention mechanisms, have made significant strides in recent years in the domain of few-shot classification. This paper commences with an introduction to the background and challenges of few-shot classification, along with a description of the principles and structure of the Transformer model. Subsequently, the paper categorizes Transformer-based few-shot image classification methods into meta-learning-based, metric-learning-based, fine-tuning-based, and feature-enhancement-based approaches, whose theoretical foundations of each method are expounded and the comparative analysis of representative algorithms are also provided. Furthermore, the paper delves into prospective research directions in this field.
Does progress on imageNet transfer to real-world datasets? We investigate this question by evaluating imageNet pre-trained models with varying accuracy (57% -83%) on six practical image classification datasets. In par...
详细信息
ISBN:
(纸本)9781713899921
Does progress on imageNet transfer to real-world datasets? We investigate this question by evaluating imageNet pre-trained models with varying accuracy (57% -83%) on six practical image classification datasets. In particular, we study datasets collected with the goal of solving real-world tasks (e.g., classifying images from camera traps or satellites), as opposed to web-scraped benchmarks collected for comparing models. On multiple datasets, models with higher imageNet accuracy do not consistently yield performance improvements. For certain tasks, interventions such as data augmentation improve performance even when architectures do not. We hope that future benchmarks will include more diverse datasets to encourage a more comprehensive approach to improving learning algorithms.
Accurate recognition of intra-pulse modulation patterns is essential for enhancing radar system performance. Tranditional recognition algorithms are typically designed under ideal conditions and handcrafted features, ...
详细信息
Because of the problem of insufficient parameter fitting accuracy of superquadric modeling method based on Levenberg-Marquardt (LM) algorithm, a superquadric modeling method of 3D object based on invasive weed optimiz...
详细信息
Face morphing attacks have posed severe threats to Face Recognition systems (FRS), which are operated in border control and passport issuance use cases. Correspondingly, morphing attack detection algorithms (MAD) are ...
详细信息
ISBN:
(纸本)9798350365474
Face morphing attacks have posed severe threats to Face Recognition systems (FRS), which are operated in border control and passport issuance use cases. Correspondingly, morphing attack detection algorithms (MAD) are needed to defend against such attacks. MAD approaches must be robust enough to handle unknown attacks in an open-set scenario where attacks can originate from various morphing generation algorithms, post-processing and the diversity of printers/scanners. The problem of generalization is further pronounced when the detection has to be made on a single suspected image. In this paper, we propose a generalized single-image-based MAD (S-MAD) algorithm by learning the encoding from Vision Transformer (ViT) architecture. Compared to CNN-based architectures, ViT model has the advantage on integrating local and global information and hence can be suitable to detect the morphing traces widely distributed among the face region. Extensive experiments are carried out on face morphing datasets generated using publicly available FRGC face datasets. Several state-of-the-art (SOTA) MAD algorithms, including representative ones that have been publicly evaluated, have been selected and benchmarked with our ViT-based approach. Obtained results demonstrate the improved detection performance of the proposed S-MAD method on inter-dataset testing (when different data is used for training and testing) and comparable performance on intra-dataset testing (when the same data is used for training and testing) experimental protocol.
This paper presents a novel image encryption algorithm that leverages the chaotic properties of the Chen system, the cryptographic strength of OpenSSL, and the mathematical robustness of the Fibonacci Q-Matrix. The pr...
详细信息
Interactive information fault diagnosis technology is a new type of fault diagnosis technology which is integrated by information fusion, artificial intelligence, computer science and other disciplines. It can extract...
详细信息
To enhance the accuracy of existing algorithms in the task of retinal vessel image segmentation, this paper proposes the incorporation of two modified convolutional blocks, in lieu of traditional ones, within the fram...
详细信息
In response to the validation needs of the directional gamma measurement system, which require going underground or using equipment containing radiation sources, and the difficulty of conducting quantitative and quali...
详细信息
imageprocessing pipelines are ubiquitous and we rely on them either directly, by filtering or adjusting an image post-capture, or indirectly, as image signal processing (ISP) pipelines on broadly deployed camera syst...
详细信息
imageprocessing pipelines are ubiquitous and we rely on them either directly, by filtering or adjusting an image post-capture, or indirectly, as image signal processing (ISP) pipelines on broadly deployed camera systems. Used by artists, photographers, system engineers, and for downstream vision tasks, traditional imageprocessing pipelines feature complex algorithmic branches developed over decades. Recently, image-to-image networks have made great strides in imageprocessing, style transfer, and semantic understanding. The differentiable nature of these networks allows them to fit a large corpus of data;however, they do not allow for intuitive, fine-grained controls that photographers find in modern photo-finishing tools. This work closes that gap and presents an approach to making complex photo-finishing pipelines differentiable, allowing legacy algorithms to be trained akin to neural networks using first-order optimization methods. By concatenating tailored network proxy models of individual processing steps (e.g. white-balance, tone-mapping, color tuning), we can model a non-differentiable reference image finishing pipeline more faithfully than existing proxy image-to-image network models. We validate the method for several diverse applications, including photo and video style transfer, slider regression for commercial camera ISPs, photography-driven neural demosaicking, and adversarial photo-editing.
暂无评论