Panoramic mosaics methods based on an 8-parameter planar homography matrix have to overcome the accumulated errors when a sequence of images loops back on itself. Usual methods are computationally intensive, and canno...
详细信息
ISBN:
(纸本)9628576623
Panoramic mosaics methods based on an 8-parameter planar homography matrix have to overcome the accumulated errors when a sequence of images loops back on itself. Usual methods are computationally intensive, and cannot ensure complete consistency of homographies. The paper presents a simple method which does not require the consistency of homographies. The method mainly exploits an un-calibrated image perspective interpolation technique (B. Yuan et al., 1998). It is therefore simple to calculate and easy to implement.
The Rapid Prototyping of Application Specific signal Processors (RASSP) program is a multi-year DARPA/Tri-Service initiative intended to dramatically improve the process by which complex digital systems, particularly ...
The Rapid Prototyping of Application Specific signal Processors (RASSP) program is a multi-year DARPA/Tri-Service initiative intended to dramatically improve the process by which complex digital systems, particularly embedded digital signal processors, are designed, manufactured, upgraded, and supported. This paper reviews the genesis of the RASSP program, considering both the problems that defined the need for the program, and the historical conditions under which it began. The RASSP program is then presented from two viewpoints. The first is programmatic, covering the goals and constraints of the program, and describing the roles of the various program participants. The second is technical, covering the major concepts upon which the developing RASSP approach to design is based and showing how the detailed technical discussions contained in the other papers in this issue of the Journal of VLSI signalprocessing relate to one another and fit into an overall development concept. The paper closes with a review of the status of the program as of this writing (Summer 1996).
It has been shown that deep recommendation models are susceptible to adversarial attacks, with this vulnerability potentially leading to significant economic losses in the e-commerce field. However, the robustness of ...
详细信息
It has been shown that deep recommendation models are susceptible to adversarial attacks, with this vulnerability potentially leading to significant economic losses in the e-commerce field. However, the robustness of deep recommendation models in response to adversarial attacks has not been systematically investigated. In this paper, therefore, we comprehensively evaluate the adversarial robustness of various representative deep models in different settings, aiming to analyze their performance impact under adversarial attacks and compare it with traditional collaborative filtering models. Notably, we examine poisoning attacks under different proportions of fake users and various popularity conditions to understand why certain deep recommendation models perform exceptionally or sub-optimally. On this basis, we further proposed practical robustness improvement strategy for the problems found in the evaluation and fully verified it through rigorous experiments. key findings include: 1) the sparser the training dataset, the weaker the robustness of a recommendation model's performance under adversarial attacks; 2) deep recommendation models exhibit greater robustness in recommending popular items under adversarial attacks, while they are more vulnerable when attacked with non-popular items; 3) the robustness of deep recommendation models is not consistently weaker than that of traditional collaborative filtering models across all attack settings. These findings highlight the security concerns in deep recommendation systems and contribute to developing more reliable models.
The Anchor-based Multi-view Subspace Clustering (AMSC) has turned into a favourable tool for large-scale multi-view clustering. However, there still exist some limitations to the current AMSC approaches. First, they t...
详细信息
The Anchor-based Multi-view Subspace Clustering (AMSC) has turned into a favourable tool for large-scale multi-view clustering. However, there still exist some limitations to the current AMSC approaches. First, they typically recover anchor graph structure in the original linear space, restricting their feasibility for nonlinear scenarios. Second, they usually overlook the potential benefits of jointly capturing the inter-view and intra-view information for enhancing the anchor representation learning. Third, these approaches mostly perform anchor-based subspace learning by a specific matrix norm, neglecting the latent high-order correlation across different views. To overcome these limitations, this paper presents an efficient and effective approach termed Large-scale Tensorized Multi-view Kernel Subspace Clustering (LTKMSC). Different from the existing AMSC approaches, our LTKMSC approach exploits both inter-view and intra-view awareness for anchor-based representation building. Concretely, the low-rank tensor learning is leveraged to capture the high-order correlation (i.e., the inter-view complementary information) among distinct views, upon which the \(l_{1,2}\) norm is imposed to explore the intra-view anchor graph structure in each view. Moreover, the kernel learning technique is leveraged to explore the nonlinear anchor-sample relationships embedded in multiple views. With the unified objective function formulated, an efficient optimization algorithm that enjoys low computational complexity is further designed. Extensive experiments on a variety of multi-view datasets have confirmed the efficiency and effectiveness of our approach when compared with the other competitive approaches.
暂无评论