Suchergebnisse - Computing methodologies Computer graphics Graphics systems AND interfaces Graphics process

StocHD: Stochastic Hyperdimensional System for Efficient and Robust Learning from Raw Data von Poduval, Prathyush, Zou, Zhuowen, Najafi, Hassan, Homayoun, Houman, Imani, Mohsen

“… Hyperdimensional Computing (HDC) is a neurally-inspired computation model working based on the observation that the human brain operates on high-dimensional representations of data, called hypervector …”
Volltext

MHDiff: Memory- and Hardware-Efficient Diffusion Acceleration via Focal Pixel Aware Quantization von Qi, Chunyu, Wang, Xuhang, Chen, Ruiyang, Yao, Yuanzheng, Jing, Naifeng, Zhang, Chen, Wang, Jun, Fu, Zhihui, Liang, Xiaoyao, Song, Zhuoran

“… Diffusion models have demonstrated superior performance in image generation tasks, thus becoming the mainstream model for generative visual tasks. Diffusion …”
Volltext

BEVSA: A Real-Time Bird's-Eye-View Semantic Segmentation Accelerator for Multi-Camera System von Lee, Sangho, Jung, Jueung, Jang, Wuyoung, Hwang, Jihyeon, Lee, Kyuho

“… A bird's-eye-view (BEV) semantic segmentation accelerator (BEVSA) is proposed for real-time 3D space perception in multi-camera system (MCS …”
Volltext

A Framework for Fine-Grained Synchronization of Dependent GPU Kernels von Jangda, Abhinav, Maleki, Saeed, Dehnavi, Maryam Mehri, Musuvathi, Madan, Saarikivi, Olli

“… These computations are commonly executed on Graphics Processing Units (GPUs), by dividing the computation into independent processing blocks, known as tiles …”
Volltext

VISTA: Optimizing GPU Scheduling through Versatile Locality-Aware Data Sharing von Falahati, Hajar, Mahani, Negin, Cristal, Adrian, Unsal, Osman

“… Graphics Processing Units (GPUs) play a pivotal role in high-performance computing by utilizing the massive parallelism of concurrent thread execution to enhance processing efficiency, known as thread-level parallelism (TLP …”
Volltext

Efficient Edge Vision Transformer Accelerator with Decoupled Chunk Attention and Hybrid Computing-In-Memory von Li, Yi, Ye, Zijian, Fu, Xiangqu, Wang, Songqi, Du, Shucheng, Lin, Ning, Shang, Dashan, Yue, Jinshan, Wang, Zhongrui, Qi, Xiaojuan, Zhang, Feng, Wang, Han

“… Vision Transformers (ViTs) are new foundation models for vision applications. Edge-deploying ViTs to realize energy-saving, low-latency, and high-performance …”
Volltext

GEM: GPU-Accelerated Emulator-Inspired RTL Simulation von Guo, Zizheng, Zhang, Yanqing, Wang, Runsheng, Lin, Yibo, Ren, Haoxing

“… ) architecture, designed for efficient CUDA execution. We also design a flow that maps circuit logic to the architecture in a process analogous to the FPGA CAD flow …”
Volltext

RADiT: Redundancy-Aware Diffusion Transformer Acceleration Leveraging Timestep Similarity von Park, Youngjun, Kim, Sangyeon, Kim, Yeonggeon, Ji, Gisan, Ryu, Sungju

“… However, a large amount of computations on the inference process and iterative sampling steps in the DiT models result in high computational costs, leading to substantial latency and energy consumption challenges …”
Volltext

Heterogeneous Acceleration Pipeline for Recommendation System Training von Adnan, Muhammad, Maboud, Yassaman Ebrahimzadeh, Mahajan, Divya, Nair, Prashant J.

“… Recommendation models rely on deep learning networks and large embedding tables, resulting in computationally and memory-intensive processes …”
Volltext

VReqST: A Requirement Specification Tool for Virtual Reality Software Products von Karre, Sai Anirudh, Halhalli, Amogha A, Reddy, Y. Raghu

“… Slight changes to the requirements can significantly impact the overall flow of the VR software development process …”
Volltext

Espresso: Exploiting the Sparsity Property in Event Sensors with Spatiotemporal Ordering von Li, Leshan, Li, Hongyi, Yang, Qingyuan, Ou, Mingtao, Zhao, Rong, Ji, Xinglong

“… However, it is challenging to efficiently stream process event data without demolishing its sparsity …”
Volltext

Towards domain-specific modelling environments based on augmented reality von Brunschwig, Léa, Campos-López, Rubén, Guerra, Esther, de Lara, Juan

“… Models are pervasive in many disciplines, like software and systems engineering. Modelling by the use of domain-specific languages (DSLs …”
Volltext

SEER: A Time Prediction Model for CNNs from GPU Kernel's View von Liu, Guodong, Wang, Sa, Bao, Yungang

“… When performing optimizations, prior knowledge about execution time of each part of the network can help avoid repeatedly time-consuming testing and profiling process …”
Volltext

Late Breaking Results: Scalable GPU-Friendly Parallelization for Sweep-Based Maze Routing von Chiang, Cheng-Yu, Cai, Zong-Ying, Lan, Chao-Chi, Chen, Yan-Jen, Hsu, Yang, Chang, Yao-Wen, Chen, Hung-Ming

“… We introduce a sweep-sharing technique that allows multiple nets to be routed simultaneously within a single sweeping process, substantially reducing memory consumption and kernel launching …”
Volltext

nvshare: Practical GPU Sharing Without Memory Size Constraints von Alexopoulos, Georgios, Mitropoulos, Dimitris

“… GPUs are essential for accelerating Machine Learning (ML) work-loads. A common practice is deploying ML jobs as containers managed by an orchestrator such as …”
Volltext

Non‐linear Rough 2D Animation using Transient Embeddings von Even, Melvin, Bénard, Pierre, Barla, Pascal

“… Traditional 2D animation requires time and dedication since tens of thousands of frames need to be drawn by hand for a typical production. Many computer …”
Volltext

TiPU: A Spatial-Locality-Aware Near-Memory Tile Processing Unit for 3D Point Cloud Neural Network von Zheng, Jiapei, Jiang, Hao, Nie, Xinkai, Huang, Zhangcheng, Chen, Chixiao, Liu, Qi

“… This paper proposes TiPU, a spatial-locality-aware near-memory tile processing unit where the point clouds are partitioned into tiles to process spatial features locally …”
Volltext

StratomeX: Visual Analysis of Large-Scale Heterogeneous Genomics Data for Cancer Subtype Characterization von Lex, A., Streit, M., Schulz, H.-J., Partl, C., Schmalstieg, D., Park, P.J., Gehlenborg, N.

“… Since there are no tools supporting this process, much of this work is done using ad‐hoc scripts and static plots, which is inefficient and limits visual exploration of the data …”
Volltext

Material Surface Reproduction and Perceptual Deformation with Projection Mapping for Car Interior Design von Takezawa, Takuro, Iwai, Daisuke, Sato, Kosuke, Hara, Toshihiro, Takeda, Yusaku, Murase, Kenji

“… Car interior design, such as dashboard, broadly consists of two parts. One is shape design, where the processes of 2D drawing, 3D modeling and evaluation with full-scale mockups are iterated, which takes a massive amount of time and cost …”
Volltext

BIM-based Mixed Reality Application for Supervision of Construction von Raimbaud, Pierre, Ruding Lou, Merienne, Frederic, Danglade, Florence, Figueroa, Pablo, Tiberio Hernandez, Jose

“… Building Information Modelling (BIM) is an up-and-coming methodology and technology used in the Architecture, Engineering and Construction (AEC …”
Volltext