Hawkeye: A Point Cloud Neural Network Processor With Virtual Pillar and Quadtree-Based Workload Management for Real-Time Outdoor BEV Detection

Large-scale 3-D processing using the point cloud neural network (PNN) has become essential for applications such as an autonomous driving system. Among the various methods, pillar-based PNN is widely utilized for processing large-scale outdoor environments. However, due to the irregular and sparse n...

Full description

Saved in:

Bibliographic Details
Published in:	IEEE journal of solid-state circuits 2025-03, Vol.60 (3), p.990-1001
Main Authors:	Lim, Sukbin, Heo, Jaehoon, Yang, Jinho, Kim, Joo-Young
Format:	Article
Language:	English
Subjects:	3-D processing Architecture Autonomous vehicles bird’s-eye-view (BEV) detection Computational efficiency convolutional neural network (CNN) Convolutional neural networks deep neural network (DNN) Efficiency Encoding Feature extraction Frames per second Graphics processing units Image processing systems Laser radar Microprocessors Neural networks neural processing unit (NPU) point cloud Point cloud compression point cloud neural network (PNN) slice-level sparsity exploitation System-on-chip Three dimensional models Three-dimensional displays Workload Workloads
Citations:	Items that this one cites
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!