scholarly journals Variational Bayesian Approach to Condition-Invariant Feature Extraction for Visual Place Recognition

2021 ◽  
Vol 11 (19) ◽  
pp. 8976
Author(s):  
Junghyun Oh ◽  
Gyuho Eoh
Keyword(s):  
Large Scale ◽  
Environmental Changes ◽  
Place Recognition ◽  
Stochastic Variational Inference ◽  
Invariant Features ◽  
Variational Autoencoder ◽  
Model Condition ◽  
Visual Place Recognition ◽  
Perceptual Changes

As mobile robots perform long-term operations in large-scale environments, coping with perceptual changes becomes an important issue recently. This paper introduces a stochastic variational inference and learning architecture that can extract condition-invariant features for visual place recognition in a changing environment. Under the assumption that a latent representation of the variational autoencoder can be divided into condition-invariant and condition-sensitive features, a new structure of the variation autoencoder is proposed and a variational lower bound is derived to train the model. After training the model, condition-invariant features are extracted from test images to calculate the similarity matrix, and the places can be recognized even in severe environmental changes. Experiments were conducted to verify the proposed method, and the experimental results showed that our assumption was reasonable and effective in recognizing places in changing environments.

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

2019 ◽  
Vol 33 ◽  
pp. 8034-8041 ◽  
Author(s):  
Kai Liu ◽  
Hua Wang ◽  
Fei Han ◽  
Hao Zhang
Keyword(s):  
Large Scale ◽  
Optimization Problem ◽  
Data Sets ◽  
Place Recognition ◽  
Benchmark Data ◽  
Localization And Mapping ◽  
Novel Method ◽  
Visual Place Recognition ◽  
Holistic Representation

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.

scholarly journals Tree-based indexing for real-time ConvNet landmark-based visual place recognition

2017 ◽  
Vol 14 (1) ◽  
pp. 172988141668695 ◽  
Author(s):  
Yi Hou ◽  
Hong Zhang ◽  
Shilin Zhou
Keyword(s):  
Real Time ◽  
Environmental Changes ◽  
Dimensional Space ◽  
Search Time ◽  
Lookup Table ◽  
Current View ◽  
Place Recognition ◽  
Linear Search ◽  
Query Image ◽  
Visual Place Recognition

Recent impressive studies on using ConvNet landmarks for visual place recognition take an approach that involves three steps: (a) detection of landmarks, (b) description of the landmarks by ConvNet features using a convolutional neural network, and (c) matching of the landmarks in the current view with those in the database views. Such an approach has been shown to achieve the state-of-the-art accuracy even under significant viewpoint and environmental changes. However, the computational burden in step (c) significantly prevents this approach from being applied in practice, due to the complexity of linear search in high-dimensional space of the ConvNet features. In this article, we propose two simple and efficient search methods to tackle this issue. Both methods are built upon tree-based indexing. Given a set of ConvNet features of a query image, the first method directly searches the features’ approximate nearest neighbors in a tree structure that is constructed from ConvNet features of database images. The database images are voted on by features in the query image, according to a lookup table which maps each ConvNet feature to its corresponding database image. The database image with the highest vote is considered the solution. Our second method uses a coarse-to-fine procedure: the coarse step uses the first method to coarsely find the top- N database images, and the fine step performs a linear search in Hamming space of the hash codes of the ConvNet features to determine the best match. Experimental results demonstrate that our methods achieve real-time search performance on five data sets with different sizes and various conditions. Most notably, by achieving an average search time of 0.035 seconds/query, our second method improves the matching efficiency by the three orders of magnitude over a linear search baseline on a database with 20,688 images, with negligible loss in place recognition accuracy.

scholarly journals NYU-VPR: Long-Term Visual Place Recognition Benchmark with View Direction and Data Anonymization Influences

2021 ◽  
Author(s):  
Diwei Sheng ◽  
Yuxiang Chai ◽  
Xinru Li ◽  
Chen Feng ◽  
Jianzhe Lin ◽  
...  
Keyword(s):  
Place Recognition ◽  
Data Anonymization ◽  
Visual Place Recognition

Multiscale Convolutional Descriptor Aggregation for Visual Place Recognition

2020 ◽  
Vol 2020 (10) ◽  
pp. 313-1-313-7
Author(s):  
Raffaele Imbriaco ◽  
Egor Bondarev ◽  
Peter H.N. de With
Keyword(s):  
Image Matching ◽  
Large Scale ◽  
Place Recognition ◽  
Aggregation Method ◽  
Local Descriptors ◽  
Multi Scale ◽  
Vector Construction ◽  
Visual Place Recognition ◽  
Different Sources ◽  
Geometric Verification

Visual place recognition using query and database images from different sources remains a challenging task in computer vision. Our method exploits global descriptors for efficient image matching and local descriptors for geometric verification. We present a novel, multi-scale aggregation method for local convolutional descriptors, using memory vector construction for efficient aggregation. The method enables to find preliminary set of image candidate matches and remove visually similar but erroneous candidates. We deploy the multi-scale aggregation for visual place recognition on 3 large-scale datasets. We obtain a Recall@10 larger than 94% for the Pittsburgh dataset, outperforming other popular convolutional descriptors used in image retrieval and place recognition. Additionally, we provide a comparison for these descriptors on a more challenging dataset containing query and database images obtained from different sources, achieving over 77% Recall@10.

SRAL: Shared Representative Appearance Learning for Long-Term Visual Place Recognition

2017 ◽  
Vol 2 (2) ◽  
pp. 1172-1179 ◽  
Author(s):  
Fei Han ◽  
Xue Yang ◽  
Yiming Deng ◽  
Mark Rentschler ◽  
Dejun Yang ◽  
...  
Keyword(s):  
Place Recognition ◽  
Visual Place Recognition

scholarly journals Towards a Robust Visual Place Recognition in Large-Scale vSLAM Scenarios Based on a Deep Distance Learning

Sensors ◽  
2021 ◽  
Vol 21 (1) ◽  
pp. 310
Author(s):  
Liang Chen ◽  
Sheng Jin ◽  
Zhoujun Xia
Keyword(s):  
Distance Learning ◽  
Large Scale ◽  
Critical Role ◽  
Distance Constraint ◽  
Place Recognition ◽  
Learning Method ◽  
Localization And Mapping ◽  
Depth Analysis ◽  
Visual Place Recognition ◽  
Distance Relationship

The application of deep learning is blooming in the field of visual place recognition, which plays a critical role in visual Simultaneous Localization and Mapping (vSLAM) applications. The use of convolutional neural networks (CNNs) achieve better performance than handcrafted feature descriptors. However, visual place recognition is still a challenging task due to two major problems, i.e., perceptual aliasing and perceptual variability. Therefore, designing a customized distance learning method to express the intrinsic distance constraints in the large-scale vSLAM scenarios is of great importance. Traditional deep distance learning methods usually use the triplet loss which requires the mining of anchor images. This may, however, result in very tedious inefficient training and anomalous distance relationships. In this paper, a novel deep distance learning framework for visual place recognition is proposed. Through in-depth analysis of the multiple constraints of the distance relationship in the visual place recognition problem, the multi-constraint loss function is proposed to optimize the distance constraint relationships in the Euclidean space. The new framework can support any kind of CNN such as AlexNet, VGGNet and other user-defined networks to extract more distinguishing features. We have compared the results with the traditional deep distance learning method, and the results show that the proposed method can improve the performance by 19–28%. Additionally, compared to some contemporary visual place recognition techniques, the proposed method can improve the performance by 40%/36% and 27%/24% in average on VGGNet/AlexNet using the New College and the TUM datasets, respectively. It’s verified the method is capable to handle appearance changes in complex environments.

scholarly journals Effects of Waves and Sediment Disturbance on Seed Bank Persistence at Tidal Flats

2021 ◽  
Vol 8 ◽  
Author(s):  
Zhenchang Zhu ◽  
Tjeerd J. Bouma ◽  
Qin Zhu ◽  
Yanpeng Cai ◽  
Zhifeng Yang
Keyword(s):  
Climate Change ◽  
Seed Bank ◽  
Large Scale ◽  
Environmental Changes ◽  
Tidal Flats ◽  
Coastal Vegetation ◽  
Climate Change Effects ◽  
Sediment Disturbance ◽  
Seed Bank Dynamics

Coastal wetlands such as salt marshes have been increasingly valued for their capacity to buffer global climate change effects, yet their long-term persistence is threatened by environmental changes. Whereas, previous studies largely focused on lateral erosion risk induced by stressors like sea level rise, it remains poorly understood of the response of lateral expansion to changing environments. Seedling establishment is a key process governing lateral marsh expansion as seen in many coastal regions such as Europe and East Asia. Here, we evaluate mechanistically the response of seed bank dynamics to changing physical disturbance at tidal flats, using the globally common coastal foundation plant, cordgrass as a model. We conducted a large-scale field study in an estuary in Northwest Europe, where seed bank dynamics of cordgrass in the tidal flats was determined and linked to in situ hydrodynamics and sediment dynamics. The results revealed that wave disturbance reduced the persistence of seeds on the surface, whereas amplified sediment disturbance lowered the persistence of both surface and buried seeds. Overall, this indicates that increasing storminess and associated sediment variability under climate change threatens seed bank persistence in tidal flats, and hence need urgently be incorporated into models for long-term bio-geomorphological development of vegetated coastal ecosystems. The knowledge gained here provides a basis for more accurate predictions on how climatically driven environmental changes may alter the fitness, resilience and persistence of coastal foundation plants, with significant implications for nature-based solutions with coastal vegetation to mitigate climate change effects.

Long-Term Visual Robot Localization using Computational Models of the Neocortex

2020 ◽  
Author(s):  
Carlos Alexandre P. Pizzino ◽  
Patricia A. Vargas ◽  
Ramon R. Costa
Keyword(s):  
Computational Models ◽  
Internal Representation ◽  
Fixed Number ◽  
Place Recognition ◽  
Biologically Inspired ◽  
Hierarchical Temporal Memory ◽  
Representational Capacity ◽  
Visual Place Recognition ◽  
High Degree

Visual place recognition is an essential capability for autonomous mobile robots which use cameras as their primary sensors. Although there has been a considerable amount of research in the topic, the high degree of image variability poses extra research challenges. Following advances in neuroscience, new biologically inspired models have been developed. Inspired by the human neocortex, hierarchical temporal memory model has potential to identify temporal sequences of spatial patterns using sparse distributed representations, which are known to have high representational capacity and high tolerance to noise. These features are interesting for place recognition applications. Some authors have proposed simplifications from the original framework, such as starting from an empty set of minicolumns and increasing the number of minicolumns on demand instead of the usage of a fixed number of minicolumns whose connections adapt over time. In this paper, we investigate the usage of framework originally proposed with the aim of extending the run-time during long-term operations. Results show that the proposed architecture can encode an internal representation of the world using a fixed number of cells in order to improve system scalability.

Sign in / Sign up

Export Citation Format

Share Document