Guardado en:
Detalles Bibliográficos
Autores principales: Yu, Fenggen, Qian, Yiming, Gil-Ureta, Francisca, Jackson, Brian, Bennett, Eric, Zhang, Hao
Formato: Preprint
Publicado: 2023
Materias:
Acceso en línea:https://arxiv.org/abs/2301.10460
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866913293148356608
author Yu, Fenggen
Qian, Yiming
Gil-Ureta, Francisca
Jackson, Brian
Bennett, Eric
Zhang, Hao
author_facet Yu, Fenggen
Qian, Yiming
Gil-Ureta, Francisca
Jackson, Brian
Bennett, Eric
Zhang, Hao
contents We present the first active learning tool for fine-grained 3D part labeling, a problem which challenges even the most advanced deep learning (DL) methods due to the significant structural variations among the small and intricate parts. For the same reason, the necessary data annotation effort is tremendous, motivating approaches to minimize human involvement. Our labeling tool iteratively verifies or modifies part labels predicted by a deep neural network, with human feedback continually improving the network prediction. To effectively reduce human efforts, we develop two novel features in our tool, hierarchical and symmetry-aware active labeling. Our human-in-the-loop approach, coined HAL3D, achieves 100% accuracy (barring human errors) on any test set with pre-defined hierarchical part labels, with 80% time-saving over manual effort.
format Preprint
id arxiv_https___arxiv_org_abs_2301_10460
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle HAL3D: Hierarchical Active Learning for Fine-Grained 3D Part Labeling
Yu, Fenggen
Qian, Yiming
Gil-Ureta, Francisca
Jackson, Brian
Bennett, Eric
Zhang, Hao
Computer Vision and Pattern Recognition
We present the first active learning tool for fine-grained 3D part labeling, a problem which challenges even the most advanced deep learning (DL) methods due to the significant structural variations among the small and intricate parts. For the same reason, the necessary data annotation effort is tremendous, motivating approaches to minimize human involvement. Our labeling tool iteratively verifies or modifies part labels predicted by a deep neural network, with human feedback continually improving the network prediction. To effectively reduce human efforts, we develop two novel features in our tool, hierarchical and symmetry-aware active labeling. Our human-in-the-loop approach, coined HAL3D, achieves 100% accuracy (barring human errors) on any test set with pre-defined hierarchical part labels, with 80% time-saving over manual effort.
title HAL3D: Hierarchical Active Learning for Fine-Grained 3D Part Labeling
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2301.10460