An algorithm is proposed for tracking objects in real time. The algorithm is based on neural network implemented on GPU. Investigation and parameter optimization of the algorithm are realized. Tracking process has accelerated by 10 times and the training process has accelerated by 2 times versus to the sequential algorithm version. The maximum resolution of the frame for real-time tracking and ...

In video compression, it can reduce the redundant information efficiently by motion estimation (ME) and motion compensation (MC), and less code is used to encode as much information as possible. In this paper, I frame encoding adopts wavelet transform and set partitioning in hierarchical trees (SPIHT) algorithm; for P frames, each frame sets the reconstructed frame of its previous frame as a re...

The continuous growth of smart communities and everincreasing demand sending or storing videos, have led to consumption huge amount data. video compression techniques are solving this emerging challenge. However, H.264 standard can be considered most notable, it has proven meet problematic requirements. authors present (BPMM) as a novel efficient Intra prediction scheme. We say that the creatio...

This paper presents a scalable caching algorithm of proxy server with the finite storage size minimizing client s buffer size and constant-bit-rate channel bandwidth. Under the general video traffic condition, it is observed that the amount of decreased client s buffer size and channel bandwidth after caching a video frame depends on the relative frame position in the time axis as well as the f...

Accurate Real-Time Reconstruction of Distant Scenes Using Computer Vision: The Recursive Multi-Frame Planar Parallax Algorithm by Todd Russell Templeton Doctor of Philosophy in Engineering-Electrical Engineering and Computer Sciences University of California, Berkeley Professor S. Shankar Sastry, Chair In this dissertation, we detail the Recursive Multi-Frame Planar Parallax (RMFPP) algorithm, ...