Object detection is a fundamental ability for robots interacting within an environment. While stunningly effective, stateof-the-art deep learning methods require huge amounts of labeled images and hours of training which does not favour such scenarios. This work presents a novel pipeline resulting from integrating (Maiettini et al. in 2017 IEEE-RAS 17th international conference on humanoid robotics (Humanoids), 2017) and (Maiettini et al. in 2018 IEEE/RSJ international conference on intelligent robots and systems (IROS), 2018), which naturally trains a robot to detect novel objects in few seconds. Moreover, we report on an extended empirical evaluation of the learning method, justifying that the proposed hybrid architecture is key in leveraging powerful deep representations while maintaining fast training time of large scale Kernel methods. We validate our approach on the Pascal VOC benchmark (Everingham et al. in Int J Comput Vis 88(2): 303-338, 2010), and on a challenging robotic scenario (iCubWorld Transformations (Pasquale et al. in Rob Auton Syst 112:260-281, 2019). We address real world use-cases and show how to tune the method for different speed/accuracy trades-off. Lastly, we discuss limitations and directions for future development.
Deep Learning (DL) methods are notoriously data hungry. Their adoption in robotics is challenging due to the cost associated with data acquisition and labeling. In this paper we focus on the problem of object detection, i.e. the simultaneous localization and recognition of objects in the scene, for which various DL architectures have been proposed in the literature. We propose to use an automatic annotation procedure, which leverages on human-robot interaction and depth-based segmentation, for the acquisition and labeling of training examples. We fine-tune the Faster R-CNN [37] network with these data acquired by the robot autonomously. We measure the performance on the same dataset and investigate the generalization abilities of the network on different settings and in absence of explicit segmentation, showing good detection performance. Experiments on the iCub humanoid robot [26] show that the proposed strategy is effective and can be used to deploy deep object detection algorithms on a robot.
Latest deep learning methods for object detection provide remarkable performance, but have limits when used in robotic applications. One of the most relevant issues is the long training time, which is due to the large size and imbalance of the associated training sets, characterized by few positive and a large number of negative examples (i.e. background). Proposed approaches are based on end-to-end learning by back-propagation [22] or kernel methods trained with Hard Negatives Mining on top of deep features [8]. These solutions are effective, but prohibitively slow for on-line applications.In this paper we propose a novel pipeline for object detection that overcomes this problem and provides comparable performance, with a 60x training speedup. Our pipeline combines (i) the Region Proposal Network and the deep feature extractor from [22] to efficiently select candidate RoIs and encode them into powerful representations, with (ii) the FALKON [23] algorithm, a novel kernel-based method that allows fast training on large scale problems (millions of points). We address the size and imbalance of training data by exploiting the stochastic subsampling intrinsic into the method and a novel, fast, bootstrapping approach.We assess the effectiveness of the approach on a standard Computer Vision dataset (PASCAL VOC 2007 [5]) and demonstrate its applicability to a real robotic scenario with the iCubWorld Transformations [18] dataset.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.