Hongyuan Cai

Learn More
This paper aims at real-time in-car video analysis to detect and track vehicles ahead for safety, autodriving, and target tracing. This paper describes a comprehensive approach to localizing target vehicles in video under various environmental conditions. The extracted geometry features from the video are continuously projected onto a 1-D profile and are(More)
Image is a dominant medium for visualizing spatial environment and creating virtual access on the Internet. Where to capture images is however subjective and relies on artistic sense of photographers so far. In this paper, we will not only visualize areas with images, but also determine where the most distinct viewpoints should be located. Starting from(More)
This work aims at detecting and tracking vehicles in in-car video. Rather than enhancing shape analysis of various vehicle types and road situations, this work focuses on vehicle and background motions because they are more general than shapes and colors of cars in various road environments. Basic features are tracked stably using corners, intensity peaks,(More)
Mobile devices equipped with GPS provide powerful tools for location-aware applications. Adding views to locations further augments users’ perception of spaces. This work creates a new type of visual map suitable for access on mobile platform. It can provide views at pervasive locations in an area. We propose a systematic approach to establish scene(More)
This work aims at real-time in-car video analysis to detect several critical events in order to alarm and assist police action. Particularly, detecting a tracked or stopped vehicle is a crucial task for further examination of suspects, protecting police safety, and remote monitoring from police station. This paper describes a comprehensive approach to(More)
This work obtains shaking-free route panoramas from a vehicle borne camera. We detect the shaking and waving profile of the camera and rectify a long route panorama automatically for an ideal parallel-perspective projection. Different from the traditional flow based approach that works on entire frame by point tracking, we create a stationary blurred(More)
Numerous videos are uploaded on video websites; most of them employ several kinds of camera operations for expanding FOV, emphasizing events, and expressing cinematic effect. To generate a profile of heterogeneous types of videos, an automatic video profiling method has been proposed to include both spatial and temporal information in a 2D image scroll. In(More)