Learn More
We argue that task requirements can be the determinant in generating different results in studies on visual object recognition. We investigated priming for novel visual objects in three implicit memory tasks. A study-test design was employed in which participants first viewed line drawings of unfamiliar objects and later made different decisions about(More)
With explosive growth of visual content repository in the Internet, a visual descriptor scheme that can support queryby-capture over the wireless link becomes attractive in a number of applications. In this work, we developed a local graph spectral embedding based model for representing high dimensional visual descriptors like SURF, SIFT, and achieves good(More)
  • 1