Improving Kernel Classifiers for Object Categorization Problems

Improving Kernel Classifiers for Object Categorization Problems. Pozdnoukhov, A. & Bengio, S. In International Conference on Machine Learning, ICML, Workshop on Learning with Partially Classified Training Data, 2005.

Paper abstract bibtex

This paper presents an approach for improving the performance of kernel classifiers applied to object categorization problems. The approach is based on the use of distributions centered around each training points, which are exploited for inter-class invariant image representation with local invariant features. Furthermore, we propose an extensive use of unlabeled images for improving the SVM-based classifier.

@inproceedings{pozdnoukhov:2005:icml,
  author = {A. Pozdnoukhov and S. Bengio},
  title = {Improving Kernel Classifiers for Object Categorization Problems},
  booktitle = {International Conference on Machine Learning, {ICML}, Workshop
on Learning with Partially Classified Training Data},
  year = 2005,
  url = {publications/ps/pozdnoukhov_2005_icml.ps.gz},
  pdf = {publications/pdf/pozdnoukhov_2005_icml.pdf},
  djvu = {publications/djvu/pozdnoukhov_2005_icml.djvu},
  original = {2005/kernel_icml_workshop},
  topics = {kernel},
  web = {http://www-connex.lip6.fr/%7Eamini/ICML05/page75-79.pdf},
  abstract = {This paper presents an approach for improving the performance of kernel classifiers applied to object categorization problems. The approach is based on the use of distributions centered around each training points, which are exploited for inter-class invariant image representation with local invariant features. Furthermore, we propose an extensive use of unlabeled images for improving the SVM-based classifier.},
  categorie = {C},
}

Downloads: 0

{"_id":"tjPqrxTsXqv5HX9GF","bibbaseid":"pozdnoukhov-bengio-improvingkernelclassifiersforobjectcategorizationproblems-2005","authorIDs":[],"author_short":["Pozdnoukhov, A.","Bengio, S."],"bibdata":{"bibtype":"inproceedings","type":"inproceedings","author":[{"firstnames":["A."],"propositions":[],"lastnames":["Pozdnoukhov"],"suffixes":[]},{"firstnames":["S."],"propositions":[],"lastnames":["Bengio"],"suffixes":[]}],"title":"Improving Kernel Classifiers for Object Categorization Problems","booktitle":"International Conference on Machine Learning, ICML, Workshop on Learning with Partially Classified Training Data","year":"2005","url":"publications/ps/pozdnoukhov_2005_icml.ps.gz","pdf":"publications/pdf/pozdnoukhov_2005_icml.pdf","djvu":"publications/djvu/pozdnoukhov_2005_icml.djvu","original":"2005/kernel_icml_workshop","topics":"kernel","web":"http://www-connex.lip6.fr/%7Eamini/ICML05/page75-79.pdf","abstract":"This paper presents an approach for improving the performance of kernel classifiers applied to object categorization problems. The approach is based on the use of distributions centered around each training points, which are exploited for inter-class invariant image representation with local invariant features. Furthermore, we propose an extensive use of unlabeled images for improving the SVM-based classifier.","categorie":"C","bibtex":"@inproceedings{pozdnoukhov:2005:icml,\n author = {A. Pozdnoukhov and S. Bengio},\n title = {Improving Kernel Classifiers for Object Categorization Problems},\n booktitle = {International Conference on Machine Learning, {ICML}, Workshop\non Learning with Partially Classified Training Data},\n year = 2005,\n url = {publications/ps/pozdnoukhov_2005_icml.ps.gz},\n pdf = {publications/pdf/pozdnoukhov_2005_icml.pdf},\n djvu = {publications/djvu/pozdnoukhov_2005_icml.djvu},\n original = {2005/kernel_icml_workshop},\n topics = {kernel},\n web = {http://www-connex.lip6.fr/%7Eamini/ICML05/page75-79.pdf},\n abstract = {This paper presents an approach for improving the performance of kernel classifiers applied to object categorization problems. The approach is based on the use of distributions centered around each training points, which are exploited for inter-class invariant image representation with local invariant features. Furthermore, we propose an extensive use of unlabeled images for improving the SVM-based classifier.},\n categorie = {C},\n}\n\n","author_short":["Pozdnoukhov, A.","Bengio, S."],"key":"pozdnoukhov:2005:icml","id":"pozdnoukhov:2005:icml","bibbaseid":"pozdnoukhov-bengio-improvingkernelclassifiersforobjectcategorizationproblems-2005","role":"author","urls":{"Paper":"http://bengio.abracadoudou.com/publications/ps/pozdnoukhov_2005_icml.ps.gz"},"downloads":0},"bibtype":"inproceedings","biburl":"http://bengio.abracadoudou.com/samy.bib","creationDate":"2020-03-18T03:43:27.399Z","downloads":0,"keywords":[],"search_terms":["improving","kernel","classifiers","object","categorization","problems","pozdnoukhov","bengio"],"title":"Improving Kernel Classifiers for Object Categorization Problems","year":2005,"dataSources":["9NCW2CDr4M3s5DvNX"]}