Pillar-based Object Detection for Autonomous Driving

Autor:	Wang, Y, Fathi, A, Kundu, A, Ross, DA, Pantofaru, C, Funkhouser, T, Solomon, J
Rok vydání:	2020
Předmět:	FOS: Computer and information sciences Computer Science - Machine Learning Computer Science - Robotics Computer Vision and Pattern Recognition (cs.CV) Computer Science - Computer Vision and Pattern Recognition Robotics (cs.RO) Machine Learning (cs.LG)
Zdroj:	arXiv
DOI:	10.48550/arxiv.2007.10323
Popis:	We present a simple and flexible object detection framework optimized for autonomous driving. Building on the observation that point clouds in this application are extremely sparse, we propose a practical pillar-based approach to fix the imbalance issue caused by anchors. In particular, our algorithm incorporates a cylindrical projection into multi-view feature learning, predicts bounding box parameters per pillar rather than per point or per anchor, and includes an aligned pillar-to-point projection module to improve the final prediction. Our anchor-free approach avoids hyperparameter search associated with past methods, simplifying 3D object detection while significantly improving upon state-of-the-art. Comment: Accepted to ECCV2020
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_dedup___::5be8289cc2b21059b0dba5a0d126c3cb Zobrazit plný text záznamu