A Video Summarization Using a Genetic Algorithm
This paper presents a video summarization method based on a Genetic Algorithm employing crossover and mutation operators to search for a meaningful summary in a video search space. In order to evaluate the keyframes, the authors define a novel fitness function which is extremely expensive to optimize with traditional methods, but which is readily optimized using GA methods. They investigate both a Binary and a Decimal Genetic Algorithm. In order to test the performance of the Binary Genetic Algorithm (BGA) method, they first compare it with the Decimal Genetic Algorithm (DGA) method. The results obtained show that the BGA more quickly and easily finds the optimal results than does the DGA. Second, they compare the BGA method with other approaches.