Active epipolar imaging simultaneously illuminates and images a scene along epipolar planes. The recent devices based on this principle, such as Episcan [1] and EpiToF [2], significantly reduce the effects of indirect light transport and ambient light, resulting in live capture of high resolution 3D at longer ranges indoors and outdoors. However, these devices are designed for narrow fields of view where lens distortion can be ignored and epipolar plane/line constraints are satisfied. In this work, we extend active epipolar imaging to obtain live-omnidirectional stereo for the first time. Instead of using a 2D sensor/projector, we use a 1D sensor and 1D light sheet source that are placed in a rectified configuration. We observe that when the lens center axis and the sensor are aligned, the distortion is mostly along that axis and epipolar plane constraint remains satisfied. This allows us to use small wide-angle lenses resulting in a compact 1D Episcan. This 1D Episcan is then spun quickly to obtain an near-spherical field of view. Based on this design, we demonstrate a custom-built hand-held working prototype for live capture of omni-directional active stereo images.

This paper describes the motivation, design and implementation of a Perception Validation System (PVS), a system for measuring the outdoor perception performance of an autonomous vehicle. The PVS relies on using large amounts of real world data and ground truth information to quantify performance aspects such as the rate of false positive or false negative detections of an obstacle detection system. Our system relies on a relational database infrastructure to achieve a high degree of flexibility in the type of analyses it can support. We discuss the main steps required for going from raw data to numerical estimates describing the performance of the perception system, including the generation of ground truth information and the safe speed metric we found to be most useful for comparing the perception system's outputs to the ground truth data. We present results illustrating some of the analyses that can be completed using the Perception Validation System.

Temporal segmentation of facial gestures in spontaneous facial behavior recorded in real-world settings is an important, unsolved, and relatively unexplored problem in facial image analysis. Several issues contribute to the challenge of this task. These include non-frontal pose, moderate to large out-of-plane head motion, large variability in the temporal scale of facial gestures, and the exponential nature of possible facial action combinations. To address these challenges, we propose a two-step approach to temporally segment facial behavior. The first step uses spectral graph techniques to cluster shape and appearance features invariant to some geometric transformations. The second step groups the clusters into temporally coherent facial gestures. We evaluated this method in facial behavior recorded during face-to- face interactions. The video data were originally collected to answer substantive questions in psychology without concern for algorithm development. The method achieved moderate convergent validity with manual FACS (Facial Action Coding System) annotation. Further, when used to preprocess video for manual FACS annotation, the method significantly improves productivity, thus addressing the need for ground-truth data for facial image analysis. Moreover, we were also able to detect unusual facial behavior.