Visual Localization in the Presence of Appearance Changes Using the Partial Order Kernel. Abdollahyan, M., Cascianelli, S., Bellocchio, E., Costante, G., Ciarfuglia, T. A., Bianconi, F., Smeraldi, F., & Fravolini, M. L. In 2018 26th European Signal Processing Conference (EUSIPCO), pages 697-701, Sep., 2018. Paper doi abstract bibtex Visual localization across seasons and under varying weather and lighting conditions is a challenging task in robotics. In this paper, we present a new sequence-based approach to visual localization using the Partial Order Kernel (POKer), a convolution kernel for string comparison, that is able to handle appearance changes and is robust to speed variations. We use multiple sequence alignment to construct directed acyclic graph representations of the database image sequences, where sequences of images of the same place acquired at different times are represented as alternative paths in a graph. We then use the POKer to compute the pairwise similarities between these graphs and the query image sequences obtained in a subsequent traversal of the environment, and match the corresponding locations. We evaluated our approach on a dataset which features extreme appearance variations due to seasonal changes. The results demonstrate the effectiveness of our approach, where it achieves higher precision and recall than two state-of-the-art baseline methods.
@InProceedings{8553252,
author = {M. Abdollahyan and S. Cascianelli and E. Bellocchio and G. Costante and T. A. Ciarfuglia and F. Bianconi and F. Smeraldi and M. L. Fravolini},
booktitle = {2018 26th European Signal Processing Conference (EUSIPCO)},
title = {Visual Localization in the Presence of Appearance Changes Using the Partial Order Kernel},
year = {2018},
pages = {697-701},
abstract = {Visual localization across seasons and under varying weather and lighting conditions is a challenging task in robotics. In this paper, we present a new sequence-based approach to visual localization using the Partial Order Kernel (POKer), a convolution kernel for string comparison, that is able to handle appearance changes and is robust to speed variations. We use multiple sequence alignment to construct directed acyclic graph representations of the database image sequences, where sequences of images of the same place acquired at different times are represented as alternative paths in a graph. We then use the POKer to compute the pairwise similarities between these graphs and the query image sequences obtained in a subsequent traversal of the environment, and match the corresponding locations. We evaluated our approach on a dataset which features extreme appearance variations due to seasonal changes. The results demonstrate the effectiveness of our approach, where it achieves higher precision and recall than two state-of-the-art baseline methods.},
keywords = {directed graphs;graph theory;image sequences;object tracking;visual databases;database image sequences;POKer;query image sequences;extreme appearance variations;seasonal changes;visual localization;appearance changes;Partial Order Kernel;weather;lighting conditions;sequence-based approach;convolution kernel;multiple sequence alignment;graph representations;Image sequences;Databases;Kernel;Europe;Signal processing;Visualization;Lighting;visual localization;partial order graphs;kernel methods},
doi = {10.23919/EUSIPCO.2018.8553252},
issn = {2076-1465},
month = {Sep.},
url = {https://www.eurasip.org/proceedings/eusipco/eusipco2018/papers/1570436699.pdf},
}
Downloads: 0
{"_id":"b6fFJY4jDqrvEBpim","bibbaseid":"abdollahyan-cascianelli-bellocchio-costante-ciarfuglia-bianconi-smeraldi-fravolini-visuallocalizationinthepresenceofappearancechangesusingthepartialorderkernel-2018","authorIDs":[],"author_short":["Abdollahyan, M.","Cascianelli, S.","Bellocchio, E.","Costante, G.","Ciarfuglia, T. A.","Bianconi, F.","Smeraldi, F.","Fravolini, M. L."],"bibdata":{"bibtype":"inproceedings","type":"inproceedings","author":[{"firstnames":["M."],"propositions":[],"lastnames":["Abdollahyan"],"suffixes":[]},{"firstnames":["S."],"propositions":[],"lastnames":["Cascianelli"],"suffixes":[]},{"firstnames":["E."],"propositions":[],"lastnames":["Bellocchio"],"suffixes":[]},{"firstnames":["G."],"propositions":[],"lastnames":["Costante"],"suffixes":[]},{"firstnames":["T.","A."],"propositions":[],"lastnames":["Ciarfuglia"],"suffixes":[]},{"firstnames":["F."],"propositions":[],"lastnames":["Bianconi"],"suffixes":[]},{"firstnames":["F."],"propositions":[],"lastnames":["Smeraldi"],"suffixes":[]},{"firstnames":["M.","L."],"propositions":[],"lastnames":["Fravolini"],"suffixes":[]}],"booktitle":"2018 26th European Signal Processing Conference (EUSIPCO)","title":"Visual Localization in the Presence of Appearance Changes Using the Partial Order Kernel","year":"2018","pages":"697-701","abstract":"Visual localization across seasons and under varying weather and lighting conditions is a challenging task in robotics. In this paper, we present a new sequence-based approach to visual localization using the Partial Order Kernel (POKer), a convolution kernel for string comparison, that is able to handle appearance changes and is robust to speed variations. We use multiple sequence alignment to construct directed acyclic graph representations of the database image sequences, where sequences of images of the same place acquired at different times are represented as alternative paths in a graph. We then use the POKer to compute the pairwise similarities between these graphs and the query image sequences obtained in a subsequent traversal of the environment, and match the corresponding locations. We evaluated our approach on a dataset which features extreme appearance variations due to seasonal changes. The results demonstrate the effectiveness of our approach, where it achieves higher precision and recall than two state-of-the-art baseline methods.","keywords":"directed graphs;graph theory;image sequences;object tracking;visual databases;database image sequences;POKer;query image sequences;extreme appearance variations;seasonal changes;visual localization;appearance changes;Partial Order Kernel;weather;lighting conditions;sequence-based approach;convolution kernel;multiple sequence alignment;graph representations;Image sequences;Databases;Kernel;Europe;Signal processing;Visualization;Lighting;visual localization;partial order graphs;kernel methods","doi":"10.23919/EUSIPCO.2018.8553252","issn":"2076-1465","month":"Sep.","url":"https://www.eurasip.org/proceedings/eusipco/eusipco2018/papers/1570436699.pdf","bibtex":"@InProceedings{8553252,\n author = {M. Abdollahyan and S. Cascianelli and E. Bellocchio and G. Costante and T. A. Ciarfuglia and F. Bianconi and F. Smeraldi and M. L. Fravolini},\n booktitle = {2018 26th European Signal Processing Conference (EUSIPCO)},\n title = {Visual Localization in the Presence of Appearance Changes Using the Partial Order Kernel},\n year = {2018},\n pages = {697-701},\n abstract = {Visual localization across seasons and under varying weather and lighting conditions is a challenging task in robotics. In this paper, we present a new sequence-based approach to visual localization using the Partial Order Kernel (POKer), a convolution kernel for string comparison, that is able to handle appearance changes and is robust to speed variations. We use multiple sequence alignment to construct directed acyclic graph representations of the database image sequences, where sequences of images of the same place acquired at different times are represented as alternative paths in a graph. We then use the POKer to compute the pairwise similarities between these graphs and the query image sequences obtained in a subsequent traversal of the environment, and match the corresponding locations. We evaluated our approach on a dataset which features extreme appearance variations due to seasonal changes. The results demonstrate the effectiveness of our approach, where it achieves higher precision and recall than two state-of-the-art baseline methods.},\n keywords = {directed graphs;graph theory;image sequences;object tracking;visual databases;database image sequences;POKer;query image sequences;extreme appearance variations;seasonal changes;visual localization;appearance changes;Partial Order Kernel;weather;lighting conditions;sequence-based approach;convolution kernel;multiple sequence alignment;graph representations;Image sequences;Databases;Kernel;Europe;Signal processing;Visualization;Lighting;visual localization;partial order graphs;kernel methods},\n doi = {10.23919/EUSIPCO.2018.8553252},\n issn = {2076-1465},\n month = {Sep.},\n url = {https://www.eurasip.org/proceedings/eusipco/eusipco2018/papers/1570436699.pdf},\n}\n\n","author_short":["Abdollahyan, M.","Cascianelli, S.","Bellocchio, E.","Costante, G.","Ciarfuglia, T. A.","Bianconi, F.","Smeraldi, F.","Fravolini, M. L."],"key":"8553252","id":"8553252","bibbaseid":"abdollahyan-cascianelli-bellocchio-costante-ciarfuglia-bianconi-smeraldi-fravolini-visuallocalizationinthepresenceofappearancechangesusingthepartialorderkernel-2018","role":"author","urls":{"Paper":"https://www.eurasip.org/proceedings/eusipco/eusipco2018/papers/1570436699.pdf"},"keyword":["directed graphs;graph theory;image sequences;object tracking;visual databases;database image sequences;POKer;query image sequences;extreme appearance variations;seasonal changes;visual localization;appearance changes;Partial Order Kernel;weather;lighting conditions;sequence-based approach;convolution kernel;multiple sequence alignment;graph representations;Image sequences;Databases;Kernel;Europe;Signal processing;Visualization;Lighting;visual localization;partial order graphs;kernel methods"],"metadata":{"authorlinks":{}},"downloads":0},"bibtype":"inproceedings","biburl":"https://raw.githubusercontent.com/Roznn/EUSIPCO/main/eusipco2018url.bib","creationDate":"2021-02-13T15:38:40.352Z","downloads":0,"keywords":["directed graphs;graph theory;image sequences;object tracking;visual databases;database image sequences;poker;query image sequences;extreme appearance variations;seasonal changes;visual localization;appearance changes;partial order kernel;weather;lighting conditions;sequence-based approach;convolution kernel;multiple sequence alignment;graph representations;image sequences;databases;kernel;europe;signal processing;visualization;lighting;visual localization;partial order graphs;kernel methods"],"search_terms":["visual","localization","presence","appearance","changes","using","partial","order","kernel","abdollahyan","cascianelli","bellocchio","costante","ciarfuglia","bianconi","smeraldi","fravolini"],"title":"Visual Localization in the Presence of Appearance Changes Using the Partial Order Kernel","year":2018,"dataSources":["yiZioZximP7hphDpY","iuBeKSmaES2fHcEE9"]}