Tilke Judd

Learning to predict where humans look

For many applications in graphics, design, and human computer interaction, it is essential to understand where humans look in a scene. Where eye tracking devices are not a viable option, models of saliency can be used to predict fixation locations. Most saliency approaches are based on bottom-up computation that does not consider top-down image semantics and often does not match actual eye movements. To address this problem, we collected eye tracking data of 15 viewers on 1003 images and use this database as training and testing examples to learn a model of saliency based on low, middle and high-level image features. This large database of eye tracking data is publicly available with this paper.

Apparent ridges for line drawing

Three-dimensional shape can be drawn using a variety of feature lines, but none of the current definitions alone seem to capture all visually-relevant lines. We introduce a new definition of feature lines based on two perceptual observations. First, human perception is sensitive to the variation of shading, and since shape perception is little affected by lighting and reflectance modification, we should focus on normal variation. Second, view-dependent lines better convey smooth surfaces. From this we define view-dependent curvature as the variation of the surface normal with respect to a viewing screen plane, and apparent ridges as the loci of points that maximize a view-dependent curvature. We present a formal definition of apparent ridges and an algorithm to render line drawings of 3D meshes. We show that our apparent ridges encompass or enhance aspects of several other feature lines.

CounterActive: an interactive cookbook for the kitchen counter

We introduce CounterActive, an interactive kitchen cookbook that teaches people to cook. After describing the interactive system and the multimedia recipe schema, we discuss results of early user test and evaluation.

Fixations on low-resolution images

When an observer looks at an image, his eyes fixate on a few select points. Fixations from different observers are often consistent-observers tend to look at the same locations. We investigate how image resolution affects fixation locations and consistency across humans through an eye-tracking experiment. We showed 168 natural images and 25 pink noise images at different resolutions to 64 observers. Each image was shown at eight resolutions (height between 4 and 512 pixels) and upsampled to 860 × 1024 pixels for display. The total amount of visual information available ranged from 1/8 to 16 cycles per degree, respectively. We measure how well one observers fixations predict another observers fixations on the same image at different resolutions using the area under the receiver operating characteristic (ROC) curves as a metric. We found that: (1) Fixations from lower resolution images can predict fixations on higher resolution images. (2) Human fixations are biased toward the center for all resolutions and this bias is stronger at lower resolutions. (3) Human fixations become more consistent as resolution increases until around 16-64 pixels (1/2 to 2 cycles per degree) after which consistency remains relatively constant despite the spread of fixations away from the center. (4) Fixation consistency depends on image complexity.

Using eye-tracking to assess different image retargeting methods

Assessing media retargeting results is not a trivial issue. When resizing one image to a particular percentage of its original size, some content has to be removed, which may affect the images original meaning and/or composition. We examine the impact of the retargeting process on human fixations, by gathering eye-tracking data for a representative benchmark of retargeted images. We compute their derived saliency maps as input to a set of computational image distance metrics. When analyzing the fixations, we found that even strong artifacts may go unnoticed for areas outside the original regions of interest. We also note that the most important alterations in semantics are due to content removal. Since using an eye tracker is not always a feasible option, we additionally show how an existing model of prediction of human fixations also works sufficiently well in a retargeting context.