Adaptive Soft Encoding: A General Unsupervised Feature Aggregation Method for Place Recognition
- 23 August 2023
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 72 (00189456) , 1-15
Abstract
Place recognition is an important problem in environment perception, which can help simultaneous localization and mapping (SLAM), as well as scene understanding of mobile robots. For place recognition, feature aggregation plays an important role in the representation of environments. In this paper, an adaptive soft-encoding (ASE) method is proposed for aggregating numerous features into low-dimensional feature vectors, which includes a training phase and an encoding phase. For the training phase, the information along each dimension of the features is evaluated which is further employed to assign all dimensions into different subdimensional intervals. After that, subdimensional Gaussian mixture models (SD-GMMs) are fitted by features in the subdimensional intervals and organized into an ASE tree, of which the tree nodes hierarchically store the parameters of the SD-GMMs to reflect the distribution of the features. For the encoding phase, features are fed into the root node of the ASE tree. The weight of each node is calculated from top to bottom to generate multiple aggregated feature vectors with different description capabilities. In addition, an ASE-based framework for place recognition is proposed in which local descriptors are extracted from sensor data (e.g., images or LiDAR data), and then aggregated into global descriptors by ASE. Finally, owing to the hierarchical structure of the ASE tree, a hierarchical matching strategy of the global descriptors is designed to recognize places efficiently. Experimental results demonstrate that feature vectors aggregated by ASE have strong distinguishability, and the ASE-based place recognition method achieves good accuracy and efficiency. The code of ASE can be accessed at https://github.com/wdyiwdwd/ASE-Encoder.Keywords
Funding Information
- Natural Science Foundation of China (U21A20486, 62073178, 61873327)
- Tianjin Science Fund for Distinguished Young Scholars (20JCJQJC00140)
- Tianjin Natural Science Foundation (20JCYBJC01470)
This publication has 32 references indexed in Scilit:
- A two-level framework for place recognition with 3D LiDAR based on spatial relation graphPattern Recognition, 2021
- Visual place recognition: A survey from deep learning perspectivePattern Recognition, 2021
- Plane-Edge-SLAM: Seamless Fusion of Planes and Edges for SLAM in Indoor EnvironmentsIEEE Transactions on Automation Science and Engineering, 2020
- A Novel Approach to Image-Sequence-Based Mobile Robot Place RecognitionIEEE Transactions on Systems, Man, and Cybernetics: Systems, 2019
- Attention-based convolutional neural network and long short-term memory for short-term detection of mood disorders based on elicited speech responsesPattern Recognition, 2019
- A Novel Global Localization Approach Based on Structural Unit Encoding and Multiple Hypothesis TrackingIEEE Transactions on Instrumentation and Measurement, 2019
- PointNetVLAD: Deep Point Cloud Based Retrieval for Large-Scale Place Recognition2018
- Visual Place Recognition: A SurveyIEEE Transactions on Robotics, 2015
- Understanding bag-of-words model: a statistical frameworkInternational Journal of Machine Learning and Cybernetics, 2010
- FAB-MAP: Probabilistic Localization and Mapping in the Space of AppearanceThe International Journal of Robotics Research, 2008