Ross Messing

Learn More
We present an activity recognition feature inspired by human psychophysical performance. This feature is based on the velocity history of tracked keypoints. We present a generative mixture model for video sequences using this feature, and show that it performs comparably to local spatio-temporal features on the KTH activity recognition dataset. In addition,(More)
Can distance perception be studied using virtual reality (VR) if distances are systematically underestimated in VR head-mounted displays (HMDs)? In an experiment in which a real environment was observed through an HMD, via live video, distances, as measured by visually directed walking, were underestimated even when the perceived environment was known to be(More)
Investigations of human perception have shown that non-local spatio-temporal information is critical and often sufficient for activity recognition. However, many recent activity recognition systems have been largely based on local space-time features and statistical techniques inspired by object recognition research. We develop a new set of statistical(More)
Distance perception in virtual reality is reviewed, and investigated in a series of experiments. In experiment 1, the systematic underestimation of distance found in virtual reality displays was investigated. This investigation used a video camera mounted on a head-mounted display to compare a photorealistic ”virtual” world to monocular viewing of the real(More)
  • 1