Search Results

The proliferation of computation-intensive applications, such as autonomous driving, has urged mobile devices to alleviate their local computation pressure using external computing resources. As a promising solution, Multi-access Edge Computing (MEC) ...

This year's SIGGRAPH Asia 2024 Courses program has received a remarkable array of submissions, showcasing the increasing interest and advancements in computer graphics, artificial intelligence, and interactive technologies. From a total of 18 submissions,...

Automatically generating high-quality textures for complex scenes remains a significant challenge in computer graphics. Recent advances in text-to-texture synthesis using 2D diffusion models have yielded impressive results for individual objects but ...

Managing the level-of-detail (LOD) in architectural models is crucial yet challenging, particularly for effective representation and visualization of buildings. Traditional approaches often fail to deliver controllable detail alongside semantic ...

We present a novel topology-preserving 3D medial axis computation framework based on volumetric restricted power diagram (RPD), while preserving the medial features and geometric convergence simultaneously, for both 3D CAD and organic shapes. The ...

Digitalization of large-scale urban scenes (in particular buildings) has been a long-standing open problem, which attributes to the challenges in data acquisition, such as incomplete scene coverage, lack of semantics, low efficiency, and low reliability ...

In order to solve the problem of accurate launch of cigarette products, an cigarette smart launch system was established based on consumer big data, aiming to improve the efficiency and effect of cigarette market launch through more scientific and ...

Achieving a high-quality and fine-grained appearance representation is an important task in Multi-Object Tracking (MOT). Recent works learn responses of different locations or channels from the global feature map to enhance visible parts or weaken ...

Accurately assessing the collateral circulation is critically essential for making acute ischemic stroke treatment plans. Current automatic methods usually rely on a single-stage CNN classifier, which typically requires a huge amount of data for ...

In this study, we tackle the complex task of generating 3D human-object interactions (HOI) from textual descriptions in a zero-shot text-to-3D manner. We identify and address two key challenges: the unsatisfactory outcomes of direct text-to-3D ...

In this paper, we introduce a novel method called FRI-Net for 2D floorplan reconstruction from 3D point cloud. Existing methods typically rely on corner regression or box regression, which lack consideration for the global shapes of rooms. To ...

Polygonal mesh reconstruction of a raw point cloud is a valuable topic in the field of computer graphics and 3D vision. Especially to 3D architectural models, polygonal mesh provides concise expressions for fundamental geometric structures while ...

Multi-view crowd localization predicts the ground locations of all people in the scene. Typical methods usually estimate the crowd density maps on the ground plane first, and then obtain the crowd locations. However, existing methods’ performances ...

We present a new approach for generating 3D house wireframes with semantic enrichment using an autoregressive model. Unlike conventional generative models that independently process vertices, edges, and faces, our approach employs a unified wire-...

In recent years, Graph Neural Networks (GNNs) have made significant strides in various applications, demonstrating their potential in handling complex networked data. Simultaneously, quantum machine learning has emerged as a rapidly advancing and ...

In this paper, for overcoming the difficulty of training dataset about two random method and manual method, the K-Nearest Neighbor (KNN) method is used to detect the network intrusion. Through adjusting K-value from part dataset and overall dataset, it ...

• A modified Gurson-Tvergaard-Needleman (GTN) model has been established for predicting ductile fracture in high strain rate conditions for alloys.
• The improved GTN model predicts results for 7xxx aluminum alloys consistent with Taylor ...

The Gurson-Tvergaard-Needleman (GTN) model has been improved to extend its application for high strain rate loading and assessed by using the Taylor impact process of 7xxx aluminum alloys. The existing modification method based on independent ...

Recently, there has been a trend of evaluating the Large Language Model (LLM) quality in the flavor of LLM-as-a-Judge, namely leveraging another LLM to evaluate the current output quality. However, existing judges are proven to be biased, namely ...

Diffusion models have achieved notable success in image generation, but they remain highly vulnerable to backdoor attacks, which compromise their integrity by producing specific undesirable outputs when presented with a pre-defined trigger. In this paper,...

In this paper, we introduce a new method for the task of interaction transfer. Given an example interaction between a source object and an agent, our method can automatically infer both surface and spatial relationships for the agent and target objects ...