Scalable Active Learning for Object Detection

Deep Neural Networks trained in a fully supervised fashion are the dominant technology in perception-based autonomous driving systems. While collecting large amounts of unlabeled data is already a major undertaking, only a subset of it can be labeled by humans due to the effort needed for high-quality annotation. Therefore, finding the right data to label has become a key challenge. Active learning is a powerful technique to improve data efficiency for supervised learning methods, as it aims at selecting the smallest possible training set to reach a required performance. We have built a scalable production system for active learning in the domain of autonomous driving. In this paper, we describe the resulting high-level design, sketch some of the challenges and their solutions, present our current results at scale, and briefly describe the open problems and future directions.

[1]  Joachim Denzler,et al.  Active Learning for Deep Object Detection , 2018, VISIGRAPP.

[2]  Andreas Nürnberger,et al.  The Power of Ensembles for Active Learning in Image Classification , 2018, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[3]  Wei Liu,et al.  SSD: Single Shot MultiBox Detector , 2015, ECCV.

[4]  Dmitry P. Vetrov,et al.  Uncertainty Estimation via Stochastic Batch Normalization , 2018, ICLR.

[5]  Wei Guo,et al.  An Adaptive Supervision Framework for Active Learning in Object Detection , 2019, BMVC.

[6]  Vinay P. Namboodiri,et al.  Deep active learning for object detection , 2018, BMVC.

[7]  Silvio Savarese,et al.  Active Learning for Convolutional Neural Networks: A Core-Set Approach , 2017, ICLR.

[8]  Yarin Gal,et al.  Uncertainty in Deep Learning , 2016 .

[9]  Ali Farhadi,et al.  You Only Look Once: Unified, Real-Time Object Detection , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[10]  John Langford,et al.  Deep Batch Active Learning by Diverse, Uncertain Gradient Lower Bounds , 2019, ICLR.

[11]  J. MacQueen.  Some methods for classification and analysis of multivariate observations , 1967 .

[12]  Joost van de Weijer,et al.  Active Learning for Deep Detection Neural Networks , 2019, 2019 IEEE/CVF International Conference on Computer Vision (ICCV).

[13]  Y. C. Pati,et al.  Orthogonal matching pursuit: recursive function approximation with applications to wavelet decomposition , 1993, Proceedings of 27th Asilomar Conference on Signals, Systems and Computers.

[14]  Charles Blundell,et al.  Simple and Scalable Predictive Uncertainty Estimation using Deep Ensembles , 2016, NIPS.

[15]  Elmar Haussmann,et al.  Less is More: An Exploration of Data Redundancy with Active Dataset Subsampling , 2019, ArXiv.

[16]  Ming-Yu Liu,et al.  Localization-Aware Active Learning for Object Detection , 2018, ACCV.

[17]  Jose M. Alvarez,et al.  Large-Scale Visual Active Learning with Deep Probabilistic Ensembles , 2018, ArXiv.

[18]  Ran El-Yaniv,et al.  Bias-Reduced Uncertainty Estimation for Deep Neural Classifiers , 2018, ICLR.

[19]  Zoubin Ghahramani,et al.  Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning , 2015, ICML.

[20]  Martial Hebert,et al.  Adaptive Semantic Segmentation with a Strategic Curriculum of Proxy Labels , 2018, ArXiv.

[21]  Sergei Vassilvitskii,et al.  k-means++: the advantages of careful seeding , 2007, SODA '07.

[22]  Burr Settles,et al.  Active Learning Literature Survey , 2009 .

[23]  Jenq-Neng Hwang,et al.  Uncertainty-Based Active Learning via Sparse Modeling for Image Classification , 2019, IEEE Transactions on Image Processing.