Abstract | ||
---|---|---|
The task of automatically tracking the visual attention in dynamic visual scenes is highly challenging. To approach it, we propose a Bayesian online learning algorithm. As the visual scene changes and new objects appear, based on a mixture model, the algorithm can identify and tell visual saccades (transitions) from visual fixation clusters (regions of interest). The approach is evaluated on real-world data, collected from eye-tracking experiments in driving sessions. |
Year | DOI | Venue |
---|---|---|
2012 | 10.1145/2168556.2168617 | ETRA |
Keywords | Field | DocType |
real-world data,visual scene change,visual saccades,eye movement data,dynamic visual scene,mixture model,eye-tracking experiment,bayesian online,visual fixation cluster,new object,visual attention,bayesian model,region of interest,eye tracking,eye movement,data collection | Online learning,Computer vision,Bayesian inference,Pattern recognition,Computer science,Eye tracking,Eye movement,Artificial intelligence,Fixation (visual),Cluster analysis,Mixture model,Bayesian probability | Conference |
Citations | PageRank | References |
25 | 1.45 | 5 |
Authors | ||
4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Enkelejda Tafaj | 1 | 48 | 3.65 |
Gjergji Kasneci | 2 | 2407 | 123.08 |
Wolfgang Rosenstiel | 3 | 1462 | 212.32 |
M Bogdan | 4 | 309 | 37.70 |