MetaView: Few-shot Active Object Recognition

In robot sensing scenarios, instead of passively utilizing human captured views, an agent should be able to actively choose informative viewpoints of a 3D object as discriminative evidence to boost the recognition accuracy. This task is referred to as active object recognition. Recent works on this task rely on a massive amount of training examples to train an optimal view selection policy. But in realistic robot sensing scenarios, the large-scale training data may not exist and whether the intelligent view selection policy can be still learned from few object samples remains unclear. In this paper, we study this new problem which is extremely challenging but very meaningful in robot sensing — Few-shot Active Object Recognition, i.e., to learn view selection policies from few object samples, which has not been considered and addressed before. We solve the proposed problem by adopting the framework of meta learning and name our method “MetaView”. Extensive experiments on both category-level and instance-level classification tasks demonstrate that the proposed method can efficiently resolve issues that are hard for state-of-the-art active object recognition methods to handle, and outperform several baselines by large margins.

[1]  Joachim Denzler,et al.  Information Theoretic Sensor Data Selection for Active Object Recognition and State Estimation , 2002, IEEE Trans. Pattern Anal. Mach. Intell..

[2]  Jeffrey Dean,et al.  Distributed Representations of Words and Phrases and their Compositionality , 2013, NIPS.

[3]  Alex Graves,et al.  Recurrent Models of Visual Attention , 2014, NIPS.

[4]  Jonathan Baxter.  Theoretical Models of Learning to Learn , 2020 .

[5]  Yandong Guo,et al.  Large Scale Incremental Learning , 2019, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[6]  Alex Graves,et al.  Neural Turing Machines , 2014, ArXiv.

[7]  Kristen Grauman,et al.  Learning to Look Around: Intelligently Exploring Unseen Environments for Unknown Tasks , 2017, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[8]  Alex Graves,et al.  Asynchronous Methods for Deep Reinforcement Learning , 2016, ICML.

[9]  Santhosh K. Ramakrishnan,et al.  Emergence of exploratory look-around behaviors through active observation completion , 2019, Science Robotics.

[10]  Subhransu Maji,et al.  Multi-view Convolutional Neural Networks for 3D Shape Recognition , 2015, 2015 IEEE International Conference on Computer Vision (ICCV).

[11]  Frank P. Ferrie,et al.  Active Object Recognition: Looking for Differences , 2001, International Journal of Computer Vision.

[12]  Krista A. Ehinger,et al.  Recognizing scene viewpoint using panoramic place representation , 2012, 2012 IEEE Conference on Computer Vision and Pattern Recognition.

[13]  Sergey Levine,et al.  Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks , 2017, ICML.

[14]  Richard Pito,et al.  A Solution to the Next Best View Problem for Automated Surface Acquisition , 1999, IEEE Trans. Pattern Anal. Mach. Intell..

[15]  Lucas Paletta,et al.  Active Object Recognition in Parametric Eigenspace , 1998, BMVC.

[16]  Jonathan Baxter,et al.  Theoretical Models of Learning to Learn , 1998, Learning to Learn.

[17]  Sven J. Dickinson,et al.  Active Object Recognition Integrating Attention and Viewpoint Control , 1997, Comput. Vis. Image Underst..

[18]  Jonathan Baxter,et al.  A Model of Inductive Bias Learning , 2000, J. Artif. Intell. Res..

[19]  Kristen Grauman,et al.  End-to-End Policy Learning for Active Visual Categorization , 2019, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[20]  Jianxiong Xiao,et al.  3D ShapeNets: A deep representation for volumetric shapes , 2014, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[21]  Oriol Vinyals,et al.  Matching Networks for One Shot Learning , 2016, NIPS.

[22]  Bernt Schiele,et al.  Transinformation for active object recognition , 1998, Sixth International Conference on Computer Vision (IEEE Cat. No.98CH36271).

[23]  Lucas Paletta,et al.  Active object recognition by view integration and reinforcement learning , 2000, Robotics Auton. Syst..

[24]  Luca Antiga,et al.  Automatic differentiation in PyTorch , 2017 .

[25]  Ronald J. Williams,et al.  Simple Statistical Gradient-Following Algorithms for Connectionist Reinforcement Learning , 2004, Machine Learning.

[26]  Pietro Perona,et al.  One-shot learning of object categories , 2006, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[27]  Kristen Grauman,et al.  Look-Ahead Before You Leap: End-to-End Active Recognition by Forecasting the Effect of Motion , 2016, ECCV.

[28]  Stefan Wermter,et al.  Continual Lifelong Learning with Neural Networks: A Review , 2019, Neural Networks.

[29]  Santhosh K. Ramakrishnan,et al.  Sidekick Policy Learning for Active Visual Exploration , 2018, ECCV.

[30]  Martial Hebert,et al.  Learning to Model the Tail , 2017, NIPS.

[31]  John K. Tsotsos,et al.  Active object recognition , 1992, Proceedings 1992 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[32]  Javier R. Movellan,et al.  Deep Q-learning for Active Recognition of GERMS: Baseline performance on a standardized dataset for active learning , 2015, BMVC.