Intentional Motion On-line Learning and Prediction

Dizan Alejandro Vasquez Govea 1 Thierry Fraichard 1 Olivier Aycard 1 Christian Laugier 1
1 E-MOTION - Geometry and Probability for Motion and Action
Inria Grenoble - Rhône-Alpes, LIG - Laboratoire d'Informatique de Grenoble
Abstract : Predicting motion of humans, animals and other objects which move according to internal plans is a challenging problem. Most existing approaches operate in two stages: a) learning typical motion patterns by observing an environment and b) predicting future motion on the basis of the learned patterns. In existing techniques, learning is performed off-line, hence, it is impossible to refine the existing knowledge on the basis of the new observations obtained during the prediction phase. We propose an approach which uses Hidden Markov Models to represent motion patterns. It is different from similar approaches because it is able to learn and predict in a concurrent fashion thanks to a novel approximate learning approach, based on the Growing Neural Gas algorithm, which estimates both HMM parameters and structure. The found structure has the property of being a planar graph, thus enabling exact inference in linear time with respect to the number of states in the model. Our experiments demonstrate that the technique works in real-time, and is able to produce sound long-term predictions of people motion.
Document type :
Journal articles
Complete list of metadatas

Cited literature [43 references]  Display  Hide  Download

https://hal.inria.fr/inria-00181663
Contributor : Thierry Fraichard <>
Submitted on : Wednesday, February 27, 2008 - 12:02:56 PM
Last modification on : Monday, August 19, 2019 - 4:42:05 PM
Long-term archiving on : Tuesday, September 21, 2010 - 3:29:43 PM

File

07-vasquez-etal-mva.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : inria-00181663, version 2

Collections

Citation

Dizan Alejandro Vasquez Govea, Thierry Fraichard, Olivier Aycard, Christian Laugier. Intentional Motion On-line Learning and Prediction. Machine Vision and Applications, Springer Verlag, 2008. ⟨inria-00181663v2⟩

Share

Metrics

Record views

529

Files downloads

538