Search

3D mesh denoising is a crucial pre-processing step in many graphics applications. However, existing data-driven mesh denoising models, primarily trained on synthetic white noise, are less effective when applied to real-world meshes with the noise of complex intensities and distributions. Moreover, how to comprehensively capture information from input meshes and apply suitable denoising models for feature-preserving mesh denoising remains a critical and unresolved challenge. This paper presents a rotation-Equivariant model-based Mesh Denoising (EMD) model and a Realistic Mesh Noise Generation (RMNG) model to address these issues. Our EMD model leverages rotation-equivariant features and self-attention weights of geodesic patches for more effective feature extraction, thereby achieving SOTA denoising results. The RMNG model, based on the Generative Adversarial Networks (GANs) architecture, generates massive amounts of realistic noisy and noiseless mesh pairs data for data-driven mesh denoising model training, significantly benefiting real-world denoising tasks. To address the smooth degradation and loss of sharp edges commonly observed in captured meshes, we further introduce varying levels of Laplacian smoothing to input meshes during the paired training data generation, endowing the trained denoising model with feature recovery capabilities. Experimental results demonstrate the superior performance of our proposed method in preserving fine-grained features while removing noise on real-world captured meshes. • Introduced an EMD method using rotation-equivariant features for SOTA mesh denoising. • Developed an RMNG model to create realistic noisy-clean pairs, enhancing real-world denoising. • Proposed a method to add smoothing degradation in data generation, improving feature recovery. [ABSTRACT FROM AUTHOR]

One of the primary challenges to apply the access control policy language XACML is the performance problem of the policy decision point (PDP), particularly when the PDP experience a great number of policies. The research on improving the PDP evaluation performance is of great significance. By combining with automaton theory an efficient policy decision engine is constructed in this paper, and attribute bitmaps are established statically for each subject, resource and action attribute of policies loaded by the policy decision engine. In evaluating access requests, the policy decision engine dynamically analyzes the requests and extracts the required attribute bitmaps to enforce the AND operation. According to the result of the AND operation, the policy decision engine matches the policies rapidly and gives out an authorization decision. The time that the policy decision engine takes to complete the evaluation of one access request is within 0.5 microsecond. This method not only greatly saves the storage space of policies, but also significantly reduces the time that the PDP takes to match the policies and evaluate access requests. Comparisons of the evaluation time taken by the policy decision engine with that taken by the Sun PDP, as well as XEngine and SBA-XACML, are made under different numbers of access requests. Experimental results show that the evaluation performance of the policy decision engine has a great improvement over that of the Sun PDP, XEngine and SBA-XACML. [ABSTRACT FROM AUTHOR]