Bidirectional integrated random fields for human behaviour understanding

Citations Over TimeTop 11% of 2012 papers

Abstract

Proposed is a bidirectional integrated random fields (BIRF) model for human behaviour understanding. The traditional hidden-state conditional random fields (HCRF) and conditional random fields (CRF) are bridged by modifying the feature functions of both, which propagates sequence classification or segmentation information in-between. Consequently, the sequence classification result by HCRF and the sequence segmentation results by CRF can be utilised to supervise the decision of each other and the performance of both models will be boosted iteratively. Large-scale experiments show that the BIRF model can achieve competing performance with the state-of-the-art methods for human behaviour understanding.

Related Papers

• → Recurrent conditional random field for language understanding(2014)112 cited
• → Hybrid semi-Markov CRF for Neural Sequence Labeling(2018)56 cited
• → Upgrading CRFS to JRFS and its Benefits to Sequence Modeling and Labeling(2020)5 cited
• → Hybrid semi-Markov CRF for Neural Sequence Labeling(2018)3 cited
• → A Novel Method of Citation Sequence Labeling Based on Conditional Random Fields(2013)