Unsupervised Learning of Probabilistic Object Models (POMs) for Object Classification, Segmentation, and Recognition Using Knowledge Propagation
We present a method to learn probabilistic object models (POMs) with minimal supervision, which exploit different visual cues and perform tasks such as classification, segmentation, and recognition. We formulate this as a structure induction and learning task and our strategy is to learn and combine...
Main Authors: | , , , |
---|---|
Other Authors: | |
Format: | Article |
Language: | en_US |
Published: |
Institute of Electrical and Electronics Engineers
2010
|
Online Access: | http://hdl.handle.net/1721.1/52348 |
_version_ | 1811076257192869888 |
---|---|
author | Zhang, Hongjiang Yuille, Alan Chen, Yuanhao Zhu, Long |
author2 | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory |
author_facet | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory Zhang, Hongjiang Yuille, Alan Chen, Yuanhao Zhu, Long |
author_sort | Zhang, Hongjiang |
collection | MIT |
description | We present a method to learn probabilistic object models (POMs) with minimal supervision, which exploit different visual cues and perform tasks such as classification, segmentation, and recognition. We formulate this as a structure induction and learning task and our strategy is to learn and combine elementary POMs that make use of complementary image cues. We describe a novel structure induction procedure, which uses knowledge propagation to enable POMs to provide information to other POMs and ldquoteach themrdquo (which greatly reduces the amount of supervision required for training and speeds up the inference). In particular, we learn a POM-IP defined on interest points using weak supervision [1], [2] and use this to train a POM-mask, defined on regional features, which yields a combined POM that performs segmentation/localization. This combined model can be used to train POM-edgelets, defined on edgelets, which gives a full POM with improved performance on classification. We give detailed experimental analysis on large data sets for classification and segmentation with comparison to other methods. Inference takes five seconds while learning takes approximately four hours. In addition, we show that the full POM is invariant to scale and rotation of the object (for learning and inference) and can learn hybrid objects classes (i.e., when there are several objects and the identity of the object in each image is unknown). Finally, we show that POMs can be used to match between different objects of the same category, and hence, enable objects recognition. |
first_indexed | 2024-09-23T10:18:46Z |
format | Article |
id | mit-1721.1/52348 |
institution | Massachusetts Institute of Technology |
language | en_US |
last_indexed | 2024-09-23T10:18:46Z |
publishDate | 2010 |
publisher | Institute of Electrical and Electronics Engineers |
record_format | dspace |
spelling | mit-1721.1/523482022-09-30T20:18:49Z Unsupervised Learning of Probabilistic Object Models (POMs) for Object Classification, Segmentation, and Recognition Using Knowledge Propagation Zhang, Hongjiang Yuille, Alan Chen, Yuanhao Zhu, Long Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory Zhu, Long Zhu, Long We present a method to learn probabilistic object models (POMs) with minimal supervision, which exploit different visual cues and perform tasks such as classification, segmentation, and recognition. We formulate this as a structure induction and learning task and our strategy is to learn and combine elementary POMs that make use of complementary image cues. We describe a novel structure induction procedure, which uses knowledge propagation to enable POMs to provide information to other POMs and ldquoteach themrdquo (which greatly reduces the amount of supervision required for training and speeds up the inference). In particular, we learn a POM-IP defined on interest points using weak supervision [1], [2] and use this to train a POM-mask, defined on regional features, which yields a combined POM that performs segmentation/localization. This combined model can be used to train POM-edgelets, defined on edgelets, which gives a full POM with improved performance on classification. We give detailed experimental analysis on large data sets for classification and segmentation with comparison to other methods. Inference takes five seconds while learning takes approximately four hours. In addition, we show that the full POM is invariant to scale and rotation of the object (for learning and inference) and can learn hybrid objects classes (i.e., when there are several objects and the identity of the object in each image is unknown). Finally, we show that POMs can be used to match between different objects of the same category, and hence, enable objects recognition. 2010-03-05T17:52:52Z 2010-03-05T17:52:52Z 2009-04 2009-01 Article http://purl.org/eprint/type/JournalArticle 0162-8828 http://hdl.handle.net/1721.1/52348 Yuanhao Chen et al. “Unsupervised Learning of Probabilistic Object Models (POMs) for Object Classification, Segmentation, and Recognition Using Knowledge Propagation.” Pattern Analysis and Machine Intelligence, IEEE Transactions on 31.10 (2009): 1747-1761. © Copyright 2010 IEEE en_US http://dx.doi.org/10.1109/tpami.2009.95 IEEE Transactions on Pattern Analysis and Machine Intelligence Article is made available in accordance with the publisher's policy and may be subject to US copyright law. Please refer to the publisher's site for terms of use. application/pdf Institute of Electrical and Electronics Engineers IEEE |
spellingShingle | Zhang, Hongjiang Yuille, Alan Chen, Yuanhao Zhu, Long Unsupervised Learning of Probabilistic Object Models (POMs) for Object Classification, Segmentation, and Recognition Using Knowledge Propagation |
title | Unsupervised Learning of Probabilistic Object Models (POMs) for Object Classification, Segmentation, and Recognition Using Knowledge Propagation |
title_full | Unsupervised Learning of Probabilistic Object Models (POMs) for Object Classification, Segmentation, and Recognition Using Knowledge Propagation |
title_fullStr | Unsupervised Learning of Probabilistic Object Models (POMs) for Object Classification, Segmentation, and Recognition Using Knowledge Propagation |
title_full_unstemmed | Unsupervised Learning of Probabilistic Object Models (POMs) for Object Classification, Segmentation, and Recognition Using Knowledge Propagation |
title_short | Unsupervised Learning of Probabilistic Object Models (POMs) for Object Classification, Segmentation, and Recognition Using Knowledge Propagation |
title_sort | unsupervised learning of probabilistic object models poms for object classification segmentation and recognition using knowledge propagation |
url | http://hdl.handle.net/1721.1/52348 |
work_keys_str_mv | AT zhanghongjiang unsupervisedlearningofprobabilisticobjectmodelspomsforobjectclassificationsegmentationandrecognitionusingknowledgepropagation AT yuillealan unsupervisedlearningofprobabilisticobjectmodelspomsforobjectclassificationsegmentationandrecognitionusingknowledgepropagation AT chenyuanhao unsupervisedlearningofprobabilisticobjectmodelspomsforobjectclassificationsegmentationandrecognitionusingknowledgepropagation AT zhulong unsupervisedlearningofprobabilisticobjectmodelspomsforobjectclassificationsegmentationandrecognitionusingknowledgepropagation |