The perception of multiple objects: a connectionist approach
The perception of multiple objects: a connectionist approach
Saccadic object recognition with an active vision system
ECAI '92 Proceedings of the 10th European conference on Artificial intelligence
Modeling visual attention via selective tuning
Artificial Intelligence - Special volume on computer vision
| Hi-index | 0.00 |
We present a biologically motivated saliency map model for active perception, and its applications to locate candidate regions of interest on various real images for further high-level analysis. The model is based on the idea that an image is memorized and recognized by the way of consecutive fixations of moving eyes on the most informative image fragments. The model suggested herein guides selecting the most informative regions of an image purely based on the properties of the input image. In order to evaluate the performance of our model, we first simulated it on various color images of natural environment, then, performed psychological human test with the same test image which were inputted to the model, and finally compared the two results to verify whether the output of the model is right. Experimental results were shown and they indicate the practicality and the promise of the model in complicated vision applications.