TY - GEN
T1 - Graph based multi-modality learning
AU - Tong, Hanghang
AU - He, Jingrui
AU - Li, Mingjing
AU - Zhang, Changshui
AU - Ma, Wei-Ying
N1 - Publication details (e.g. title, author(s), publication statuses and dates) are captured on an “AS IS” and “AS AVAILABLE” basis at the time of record harvesting from the data source. Suggestions for further amendments or supplementary information can be sent to [email protected].
PY - 2005
Y1 - 2005
N2 - To better understand the content of multimedia, a lot of research efforts have been made on how to learn from multi-modal feature. In this paper, it is studied from a graph point of view: each kind of feature from one modality is represented as one independent graph; and the learning task is formulated as inferring from the constraints in every graph as well as supervision information (if available). For semi-supervised learning, two different fusion schemes, namely linear form and sequential form, are proposed. For each scheme, it is derived from optimization point of view; and further justified from two sides: similarity propagation and Bayesian interpretation. By doing so, we reveal the regular optimization nature, transductive learning nature as well as prior fusion nature of the proposed schemes, respectively. Moreover, the proposed method can be easily extended to unsupervised learning, including clustering and embedding. Systematic experimental results validate the effectiveness of the proposed method. Copyright © 2005 ACM.
AB - To better understand the content of multimedia, a lot of research efforts have been made on how to learn from multi-modal feature. In this paper, it is studied from a graph point of view: each kind of feature from one modality is represented as one independent graph; and the learning task is formulated as inferring from the constraints in every graph as well as supervision information (if available). For semi-supervised learning, two different fusion schemes, namely linear form and sequential form, are proposed. For each scheme, it is derived from optimization point of view; and further justified from two sides: similarity propagation and Bayesian interpretation. By doing so, we reveal the regular optimization nature, transductive learning nature as well as prior fusion nature of the proposed schemes, respectively. Moreover, the proposed method can be easily extended to unsupervised learning, including clustering and embedding. Systematic experimental results validate the effectiveness of the proposed method. Copyright © 2005 ACM.
KW - Bayesian interpretation
KW - Graph model
KW - Multi-modality analysis
KW - Regularized optimization
KW - Similarity propagation
UR - http://www.scopus.com/inward/record.url?scp=84883062662&partnerID=8YFLogxK
UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-84883062662&origin=recordpage
U2 - 10.1145/1101149.1101337
DO - 10.1145/1101149.1101337
M3 - RGC 32 - Refereed conference paper (with host publication)
SN - 1595930442
SN - 9781595930446
T3 - Proceedings of the 13th ACM International Conference on Multimedia, MM 2005
SP - 862
EP - 871
BT - Proceedings of the 13th ACM International Conference on Multimedia, MM 2005
T2 - 13th ACM International Conference on Multimedia, MM 2005
Y2 - 6 November 2005 through 11 November 2005
ER -