Attention-guided object and action recognition based on probabilistic learning and feature boosting for understanding human-object interaction

• Project/Area Number: 23500188
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Intelligent informatics
• Research Institution: Soka University
• Principal Investigator: ATSUMI Masayasu 創価大学, 工学部, 准教授 (00192980)
• Project Period (FY): 2011 – 2013
• Project Status: Completed (Fiscal Year 2013)
• Budget Amount: ¥5,330,000 (Direct Cost: ¥4,100,000、Indirect Cost: ¥1,230,000); Fiscal Year 2013: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000); Fiscal Year 2012: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000); Fiscal Year 2011: ¥3,250,000 (Direct Cost: ¥2,500,000、Indirect Cost: ¥750,000)
• Keywords: 注意 / 確率的学習 / ブースティング / 物体認識 / アクション認識 / コンテキスト / 意味ネットワーク / 確率的推論 / 人工知能 / 機械学習 / パターン認識 / カテゴリゼーション / コンピュータビジョン / 選択的注意 / 知能ロボティクス

Research Abstract

This research proposed probabilistic methods of attention-guided object recognition and object-oriented action recognition for understanding human-object interaction. In the proposed methods, attention-guided object recognition is performed in the context of co-occurring objects by using a classification tree which is learned based on the probabilistic latent component tree analysis and feature boosting. Also object-oriented action recognition is performed in the mutual contexts of objects and actions by using a probabilistic semantic network of visual motion classes and their semantic tags which is learned based on the incremental probabilistic latent component analysis. It was shown that the proposed method achieved high recognition accuracy through experiments using image data sets of plural object categories and also a set of video clips of object-oriented actions captured by a Kinect sensor mounted on a robot.

Research Products

• [Journal Article] Learning Probabilistic Semantic Network of Objectoriented Action and Activity (2014)
  • Author(s): Masayasu Atsumi
  • Journal Title: Lecture Notes in Artificial Intelligence
  • Peer Reviewed
• [Journal Article] Object Categorization in Context based on Probabilistic Learning of Classification Tree with Boosted Features and Co-occurrence Structure (2013)
  • Author(s): Masayasu Atsumi
  • Journal Title: Lecture Notes in Computer Science : Advances in Visual Computing Volume: Vol.8033 Pages: 416-426
  • DOI: 10.1007/978-3-642-41914-0_41
  • ISBN: 9783642419133, 9783642419140
  • Peer Reviewed
• [Journal Article] Attention-Guided Organized Perception and Learning of Object Categories Based on Probabilistic Latent Variable Models (2013)
  • Author(s): Masayasu Atsumi
  • Journal Title: Journal of Intelligent Learning Systems and Applications Volume: Vol.5,No.2 Issue: 02 Pages: 123-133
  • DOI: 10.4236/jilsa.2013.52014
  • Peer Reviewed
• [Journal Article] Attention-guided Organized Perception and Learning of Object Categories based on Probabilistic Latent Variable Models (2013)
  • Author(s): Masayasu Atsumi
  • Journal Title: Journal of Intelligent Learning Systems and Applications Volume: Vol.5, No.2
  • Peer Reviewed
• [Journal Article] Learning Visual Categories based on Probabilistic Latent Component Models with Semisupervised Labeling (2012)
  • Author(s): Masayasu Atsumi
  • Journal Title: GSTF International Journal on Computing Volume: Vol.2,No.1 Pages: 88-93
  • Peer Reviewed
• [Journal Article] Visual Categorization based on Learning Contextual Probabilistic Latent Component Tree (2012)
  • Author(s): Masayasu Atsumi
  • Journal Title: Lecture Notes in Computer Science : Artificial Neural Networks and Machine Learning Volume: Vol.7552 Pages: 419-426
  • DOI: 10.1007/978-3-642-33269-2_53
  • ISBN: 9783642332685, 9783642332692
  • Peer Reviewed
• [Journal Article] Object and Scene Recognition based on Learning Probabilistic Latent Component Tree with Boosted Features (2012)
  • Author(s): Masayasu Atsumi
  • Journal Title: International Journal of Machine Learning and Computing Volume: Vol.2,No.6 Pages: 762-766
  • DOI: 10.7763/ijmlc.2012.v2.232
  • Peer Reviewed
• [Journal Article] Learning Visual Categories based on Probabilistic Latent Component Models with Semi-supervised Labeling (2012)
  • Author(s): Masayasu Atsumi
  • Journal Title: GSTF International Journal on Computing Volume: Vol.2, No.1 Pages: 88-93
  • Peer Reviewed
• [Journal Article] Visual Category Learning based on Probabilistic Latent Component Models with Semi-supervised Labeling (2011)
  • Author(s): Masayasu Atsumi
  • Journal Title: Proceedings of 2nd Annual International Conference on Advanced Topics in Artificial Intelligence Volume: - Pages: 50-55
  • Peer Reviewed
• [Presentation] Learning Probabilistic Semantic Network of Object-oriented Action and Activity (2014)
  • Author(s): Masayasu Atsumi
  • Organizer: The 16th International Conference on Artificial Intelligence: Methodology, Systems, Applications
  • Place of Presentation: Varna, Bulgaria
• [Presentation] 物体指向動作の心象と表象の確率的カテゴリゼーション (2014)
  • Author(s): 渥美雅保
  • Organizer: 2014年度人工知能学会第28回全国大会
  • Place of Presentation: 松山市, 日本
• [Presentation] 物体指向動作の心象と表象の確率的カテゴリゼーション (2014)
  • Author(s): 渥美雅保
  • Organizer: 2014年度人工知能学会第28回全国大会
  • Place of Presentation: 松山市
• [Presentation] Object Categorization in Context based on Probabilistic Learning of Classification Tree with Boosted Features and Co-occurrence Structure (2013)
  • Author(s): Masayasu Atsumi
  • Organizer: 9th International Symposium on Visual Computing
  • Place of Presentation: Rethymnon, Crete, Greece
• [Presentation] 注意と共起に基づくシーンの学習と認識 (2013)
  • Author(s): 渥美雅保
  • Organizer: 2013年度人工知能学会第27回全国大会
  • Place of Presentation: 富山市, 日本
• [Presentation] 注意と共起に基づくシーンの学習と認識 (2013)
  • Author(s): 渥美雅保
  • Organizer: 2013年度人工知能学会第27回全国大会
  • Place of Presentation: 富山市
• [Presentation] 注意と共起に基づくシーンの学習と認識 (2013)
  • Author(s): 渥美雅保
  • Organizer: 2013年度人工知能学会全国大会
  • Place of Presentation: 日本，富山
• [Presentation] Object and Scene Recognition based on Learning Probabilistic Latent Component Tree with Boosted Features (2012)
  • Author(s): Masayasu Atsumi
  • Organizer: 2012 International Conference on Information and Intelligent Computing
  • Place of Presentation: Chengdu, China
• [Presentation] Visual Categorization based on Learning Contextual Probabilistic Latent Component Tree (2012)
  • Author(s): Masayasu Atsumi
  • Organizer: 22nd International Conference on Artificial Neural Networks
  • Place of Presentation: Lausanne, Switzerland
• [Presentation] 前景・後景コンテキストからのラベル付き物体カテゴリ木学習に基づく共起物体認識 (2012)
  • Author(s): 渥美雅保
  • Organizer: 2012年度人工知能学会第26回全国大会
  • Place of Presentation: 山口市, 日本
• [Presentation] 前景・後景コンテキストからのラベル付き物体カテゴリ木学習に基づく共起物体認識 (2012)
  • Author(s): 渥美雅保
  • Organizer: 2012年度人工知能学会全国大会
  • Place of Presentation: 日本，山口
• [Presentation] 前景・後景コンテキストからのラベル付き物体カテゴリ木学習に基づく共起物体認識 (2012)
  • Author(s): 渥美雅保
  • Organizer: 2012年度人工知能学会全国大会
  • Place of Presentation: 山口県教育会館 (山口市)
• [Presentation] Visual Category Learning based on Probabilistic Latent Component Models with Semi-supervised Labeling (2011)
  • Author(s): Masayasu Atsumi
  • Organizer: 2nd Annual International Conference on Advanced Topics in Artificial Intelligence
  • Place of Presentation: Singapore
• [Presentation] 確率潜在コンポーネント木による物体カテゴリ構成の学習 (2011)
  • Author(s): 渥美雅保
  • Organizer: 2011年度人工知能学会第25回全国大会
  • Place of Presentation: 盛岡市, 日本
• [Presentation] Visual Category Learning based on Probabilistic Latent Component Models with Semi-supervised Labeling (2011)
  • Author(s): Masayasu Atsumi
  • Organizer: 2nd Annual International Conference on Advanced Topics in Artificial Intelligence
  • Place of Presentation: Hotel Fort Canning (Singapore)
• [Presentation] 確率潜在コンポーネント木による物体カテゴリ構成の学習 (2011)
  • Author(s): 渥美雅保
  • Organizer: 2011年度人工知能学会全国大会
  • Place of Presentation: アイーナいわて県民情報交流センター (盛岡市)