arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2302.03566cs.CV
收藏

环顾与学习:基于探索的自训练目标检测

Look Around and Learn: Self-Training Object Detection by Exploration

Istituto Italiano di Tecnologia(意大利技术研究院);University of Genoa(热那亚大学)

Gianluca Scarpellini, Stefano Rosa, Pietro Morerio, Lorenzo Natale, Alessio Del Bue

AI总结:

本文提出完全自监督的具身探索框架,通过Look Around策略挖掘困难样本并利用Disagreement Reconciliation机制精炼伪标签,在模拟和真实场景中分别将目标检测性能提升6.2%和9.97%。

摘要:

当目标检测器部署在新环境中时,性能通常会下降。本文研究具身智能体如何在新环境中探索并获取图像时,自动微调预存的目标检测器,无需人工干预,即完全自监督方法。在我们的设定中,智能体最初使用预训练的现成检测器探索环境,定位目标并关联伪标签。通过假设同一目标在不同视角下的伪标签必须一致,我们学习了探索策略Look Around来挖掘困难样本,并设计了一种名为Disagreement Reconciliation的新机制,从观测的一致性中生成精炼的伪标签。我们实现了当前最先进方法的统一基准,并将我们的方法与现有探索策略和感知机制进行比较。结果表明,我们的方法优于现有方法,在模拟场景中将目标检测器提升6.2%,比其他最先进方法提升3.59%,在真实机器人测试中提升9.97%,且不依赖真实标签。所提方法和基线的代码可在https://iit-pavis.github.io/Look_Around_And_Learn/获取。

英文摘要:

When an object detector is deployed in a novel setting it often experiences a drop in performance. This paper studies how an embodied agent can automatically fine-tune a pre-existing object detector while exploring and acquiring images in a new environment without relying on human intervention, i.e., a fully self-supervised approach. In our setting, an agent initially learns to explore the environment using a pre-trained off-the-shelf detector to locate objects and associate pseudo-labels. By assuming that pseudo-labels for the same object must be consistent across different views, we learn the exploration policy Look Around to mine hard samples, and we devise a novel mechanism called Disagreement Reconciliation for producing refined pseudo-labels from the consensus among observations. We implement a unified benchmark of the current state-of-the-art and compare our approach with pre-existing exploration policies and perception mechanisms. Our method is shown to outperform existing approaches, improving the object detector by 6.2% in a simulated scenario, a 3.59% advancement over other state-of-the-art methods, and by 9.97% in the real robotic test without relying on ground-truth. Code for the proposed approach and baselines are available at https://iit-pavis.github.io/Look_Around_And_Learn/.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑