Video sequences are major sources of traffic for broadband ISDN networks, and video compression is fundamental to the efficient use of such networks. We present a novel neural method to achieve real-time adaptive compression of video. This tends to maintain a target quality of the decompressed image specified by the user. The method uses a set of compression/decompression neural networks of different levels of compression, as well as a simple motiondetection procedure. We describe the method and present experimental data concerning its performance and traffic characteristics with real video sequences. The impact of this compression method on ATM-cell traffic is also investigated and measurement data are provided.

No abstract

No abstract

Abstvact-The serializability problem, which isNP-complete, is to decide on the existence of a total order consistent with a given ordering constraint. In this paper, a connectionist machine is proposed for finding an 'almost sure' solution to the serializability problem such that a feasible final configuration implies that the set is serializable with the given ordering constraint but an infeasible final configuration does not mean that it is not serializable. Therefore, the machine is not able to recognize all the serializable constraints, however the experiments show that the machine seems to converge to a stable configuration in polynomial time and the performance of the machine is quite high, i.e. for most of the time, it is able to find the correct answer.