The aim of this study was to detect and characterize noroviruses (NoVs) in environmental water samples. One hundred and fourteen water samples were collected from a river and irrigation canals in central Thailand during 2006-2007. NoVs were detected by RT-nested PCR in 13% of the samples. The river samples (22%) contained NoVs at a higher frequency than the irrigation canal samples (4%). Among ...

We propose a method for human action recognition from still images that uses the silhouette and the upper body as a proxy for the pose of the person, and also to guide alignment between instances for the purpose of computing registered feature descriptors. Our contributions include an efficient algorithm, formulated as an energy minimization, for using the silhouette to align body parts between...

Accurate metering of agricultural water is becoming increasingly important world wide. Surprisingly, the dynamics of water in small irrigation canals are not well understood and this limits the accuracy of flow measurements in these channels. To help improve the accuracy of flow data, we performed detailed measurements of the velocity distribution for a variety of flow conditions in a several s...

We train a kernel SVM for each action class. The SVM kernel is a convex combination of base kernels, which capture different visual cues: HOG, SIFT, color, pose, object detection scores. Some of these cues are computed at various relative locations of the provided human bounding box, yielding a total of 20 kernels. We optimize the weights for kernel combination using randomized grid search. Thi...

ID No : 255