Glimpse is a continuous, real-time object recognition system for camera-equipped mobile devices. Glimpse captures full-motion video, locates objects of interest, recognizes and labels them, and tracks them from frame to frame for the user. Because the algorithms for object recognition entail significant computation, Glimpse runs them on server machines. When the latency between the server and mobile device is higher than a frame-time, this approach lowers object recognition accuracy. To regain accuracy, Glimpse uses an active cache of video frames on the mobile device. A subset of the frames in the active cache are used to track objects on the mobile, using (stale) hints about objects that arrive from the server from time to time. To reduce network bandwidth usage, Glimpse computes trigger frames to send to the server for recognizing and labeling. Experiments with Android smartphones and Google Glass over Verizon, AT&T, and a campus Wi-Fi network show that with hardware face detection support (available on many mobile devices), Glimpse achieves precision between 96.4% to 99.8% for continuous face recognition, which improves over a scheme performing hardware face detection and server-side recognition without Glimpse's techniques by between 1.8-2.5×. The improvement in precision for face recognition without hardware detection is between 1.6-5.5×. For road sign recognition, which does not have a hardware detector, Glimpse achieves precision between 75% and 80%; without Glimpse, continuous detection is non-functional (0.2%-1.9% precision).

Glimpse is a continuous, real-time object recognition system for camera-equipped mobile devices. Glimpse captures full-motion video, locates objects of interest, recognizes and labels them, and tracks them from frame to frame for the user. Because the algorithms for object recognition entail significant computation, Glimpse runs them on server machines. To achieve high accuracy, Glimpse uses an active cache of video frames on the mobile device. A subset of the frames in the active cache are used to track objects on the mobile, using (stale) hints about objects that arrive from the server from time to time. To reduce network bandwidth usage, Glimpse computes trigger frames to send to the server for recognizing and labeling.

This study presents several extensions to our previous work on the PipeProbe system, a mobile sensor system for identifying the spatial topology of hidden water pipelines (i.e., non-moldable pipes such as copper and PVC) behind walls or under floors [Lai et al. 2010]. The PipeProbe system works by dropping a tiny wireless sensor capsule into the source of a water pipeline. As the PipeProbe capsule traverses the pipelines, it gathers and transmits pressure and angular velocity readings. Through spatiotemporal analysis of these sensor readings, the proposed algorithm locates all turning points in the pipelines and maps their 3D spatial topology. This study expands upon previous research by developing new sensing techniques that identify variable-diameter pipes and differentiate 90-degree pipe turns from 45-degree pipe bends.

Smartphones now are equipped with a variety of sensors including inertial sensors (accelerometers and gyroscopes) and multiple position sensors (GPS, WiFi, and cellular radios). These powerful capabilities have made smartphones an attractive platform for tasking applications. Tasking applications are rapid developing mobile applications which process data from multiple sensors continuously to determine user's context (such as location or activity) and take certain actions based on pre-defined conditions. Examples of such applications include location-based reminders, notifying when friends are nearby, changing the ring-mode of a phone automatically depending on the location, automatically tracking and storing movement tracks when driving, and inferring the number of steps walked each day. However, today, developing tasking applications is non-trivial for two reasons: poor abstractions and poor programming support [1].

This paper presents PipeProbe, a mobile sensor system for determining the spatial topology of hidden water pipelines behind walls. PipeProbe works by dropping a tiny wireless sensor capsule into the source of the water pipelines. As the PipeProbe capsule traverses the pipelines, it gathers and transmits pressure and angular velocity readings. Through spatio-temporal analysis on the sensor readings, our algorithm locates all turning points in the pipelines and maps their 3D spatial topology. We evaluated the PipeProbe system by developing a prototype and using data collected in our experimental testbed. Results show that the PipeProbe system successfully located and estimated 90% of all pipe tube lengths within 8-cm accuracy on average tube lengths of 76 cm. PipeProbe also successfully located 90% of all turning points within 15-cm accuracy on average length paths of 335cm.

We propose PipeProbe, a mobile sensor system for mapping hidden water pipelines inside cement walls or under floor coverings. PipeProbe works by dropping a sensor capsule into the source of the water pipelines. As the PipeProbe capsule traverses the pipelines, it gathers water pressure and accelerometer readings. Given the pressure sensor readings, we can identify the height of PipeProbe. Accelerometer is used to recognize different types of water pipeline shape when PipeProbe travels through. The PipeProbe system is non-intrusive and requires no alteration to the water pipeline infrastructure.