Statistical models of object interaction

被引：26

作者：

Morris, RJ ^{[1
]}

Hogg, DC

机构：

[1] Univ Leeds, Dept Stat, Leeds LS2 9JT, W Yorkshire, England

[2] Univ Leeds, Sch Comp Studies, Leeds LS2 9JT, W Yorkshire, England

来源：

INTERNATIONAL JOURNAL OF COMPUTER VISION | 2000年 / 37卷 / 02期

基金：

英国工程与自然科学研究理事会;

关键词：

behaviour modelling; object interactions; landmark data;

D O I：

10.1023/A:1008159822101

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

We present a method for assessing the likelihood of a trajectory of an object through a scene consisting of a number of other objects. The closest points on the trajectory to the other objects are chosen as landmark points and at each landmark we calculate the probability of the interaction based on the speed and distance. Sequences of such probabilities are then sorted in increasing order. Finally a weighted sum of the first few elements in this weighted list is used to classify trajectories in a supervised learning framework.

引用

页码：209 / 215

页数：7

共 17 条

[1]

[Anonymous], 1995, P INT WORKSH AUT FAC

[2]

BAUMBERG A, 1994, P EUR C COMP VIS, V1, P299

[3]

BAUMBERG A, 1996, IMAGE VISION COMPUTI

[4]

Baumberg A. M., 1994, Proceedings of the 1994 IEEE Workshop on Motion of Non-Rigid and Articulated Objects (Cat. No.94TH0671-8), P194, DOI 10.1109/MNRAO.1994.346236

[5]

Bobick A. F., 1995, Proceedings. Fifth International Conference on Computer Vision (Cat. No.95CB35744), P382, DOI 10.1109/ICCV.1995.466914

[6]

Bookstein F L, 1997, Med Image Anal, V1, P225

[7]

Bookstein F. L., 1997, MORPHOMETRIC TOOLS L

[8] Learning and recognizing human dynamics in video sequences [J].

Bregler, C .

1997 IEEE COMPUTER SOCIETY CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, PROCEEDINGS, 1997, :568-574

[9]

BULPITT AJ, 1993, P WORLD C NEUR NETW, V3, P708

[10] VISUAL SURVEILLANCE IN A DYNAMIC AND UNCERTAIN WORLD [J].

BUXTON, H ;

GONG, SG .

ARTIFICIAL INTELLIGENCE, 1995, 78 (1-2) :431-459

← 1 2 →