Visual Place Recognition via Robust ℓ2-Norm Distance Based Holism and Landmark Integration

Visual place recognition is essential for large-scale simultaneous localization and mapping (SLAM). Long-term robot operations across different time of the days, months, and seasons introduce new challenges from significant environment appearance variations. In this paper, we propose a novel method to learn a location representation that can integrate the semantic landmarks of a place with its holistic representation. To promote the robustness of our new model against the drastic appearance variations due to long-term visual changes, we formulate our objective to use non-squared l2-norm distances, which leads to a difficult optimization problem that minimizes the ratio of the l2,1-norms of matrices. To solve our objective, we derive a new efficient iterative algorithm, whose convergence is rigorously guaranteed by theory. In addition, because our solution is strictly orthogonal, the learned location representations can have better place recognition capabilities. We evaluate the proposed method using two large-scale benchmark data sets, the CMU-VL and Nordland data sets. Experimental results have validated the effectiveness of our new method in long-term visual place recognition applications.

[1]  Hua Wang,et al.  Learning of Holism-Landmark Graph Embedding for Place Recognition in Long-Term Autonomy , 2018, IEEE Robotics and Automation Letters.

[2]  Feiping Nie,et al.  Early Active Learning via Robust Representation and Structured Sparsity , 2013, IJCAI.

[3]  Feiping Nie,et al.  Robust Principal Component Analysis with Non-Greedy l1-Norm Maximization , 2011, IJCAI.

[4]  Feiping Nie,et al.  Robust and Discriminative Self-Taught Learning , 2013, ICML.

[5]  Niko Sünderhauf,et al.  BRIEF-Gist - closing the loop by simple means , 2011, 2011 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[6]  Kai Liu,et al.  High-Order Co-Clustering via Strictly Orthogonal and Symmetric L1-Norm Nonnegative Matrix Tri-Factorization , 2018, IJCAI.

[7]  C. S. G. Lee,et al.  Robust Semantic Place Recognition with Vocabulary Tree and Landmark Detection , 2011 .

[8]  Gordon Wyeth,et al.  SeqSLAM: Visual route-based navigation for sunny summer days and stormy winter nights , 2012, 2012 IEEE International Conference on Robotics and Automation.

[9]  Junbin Gao,et al.  Robust L1 Principal Component Analysis and Its Bayesian Variational Inference , 2008, Neural Computation.

[10]  Peter I. Corke,et al.  Visual Place Recognition: A Survey , 2016, IEEE Transactions on Robotics.

[11]  Hua Wang,et al.  Robust Multimodal Sequence-Based Loop Closure Detection via Structured Sparsity , 2016, Robotics: Science and Systems.

[12]  Kai Liu,et al.  Learning Multi-instance Enriched Image Representations via Non-greedy Ratio Maximization of the l1-Norm Distances , 2018, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[13]  Kai Liu,et al.  Robust Multi-Relational Clustering via \ell_1-Norm Symmetric Nonnegative Matrix Factorization , 2015, ACL.

[14]  Feiping Nie,et al.  Multi-View Clustering and Feature Learning via Structured Sparsity , 2013, ICML.

[15]  Hyun Myung,et al.  Image feature-based real-time RGB-D 3D SLAM with GPU acceleration , 2013 .

[16]  Xue Yang,et al.  SRAL: Shared Representative Appearance Learning for Long-Term Visual Place Recognition , 2017, IEEE Robotics and Automation Letters.

[17]  John J. Leonard,et al.  Sparse optimization for robust and efficient loop closing , 2017, Robotics Auton. Syst..

[18]  John Wright,et al.  Robust Principal Component Analysis: Exact Recovery of Corrupted Low-Rank Matrices via Convex Optimization , 2009, NIPS.

[19]  Michael Milford,et al.  Place Recognition with ConvNet Landmarks: Viewpoint-Robust, Condition-Robust, Training-Free , 2015, Robotics: Science and Systems.

[20]  Wolfram Burgard,et al.  Robust visual SLAM across seasons , 2015, 2015 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[21]  Feiping Nie,et al.  Learning a Mahalanobis distance metric for data clustering and classification , 2008, Pattern Recognit..

[22]  Feiping Nie,et al.  Robust Distance Metric Learning via Simultaneous L1-Norm Minimization and Maximization , 2014, ICML.

[23]  Cindy Cappelle,et al.  Place Recognition Based Visual Localization Using LBP Feature and SVM , 2015, MICAI.

[24]  Takeo Kanade,et al.  Real-time topometric localization , 2012, 2012 IEEE International Conference on Robotics and Automation.

[25]  Niko Sünderhauf,et al.  Are We There Yet? Challenging SeqSLAM on a 3000 km Journey Across All Four Seasons , 2013 .

[26]  James M. Rehg,et al.  CENTRIST: A Visual Descriptor for Scene Categorization , 2011, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[27]  Feiping Nie,et al.  A Non-Greedy Algorithm for L1-Norm LDA , 2017, IEEE Transactions on Image Processing.

[28]  Hua Wang,et al.  Learning Integrated Holism-Landmark Representations for Long-Term Loop Closure Detection , 2018, AAAI.

[29]  Wolfram Burgard,et al.  Robust Visual Robot Localization Across Seasons Using Network Flows , 2014, AAAI.

[30]  I. Jolliffe.  Principal Component Analysis and Factor Analysis , 1986 .

[31]  Feiping Nie,et al.  Globally and Locally Consistent Unsupervised Projection , 2014, AAAI.