A Review on: Automatic Movie Character Annotation by Robust Face-Name Graph Matching

Now a day's character Identification from films is a very challenging task due to the huge variation in the appearance of each & every character. It will lead significant research interests and may have many interesting applications in today's life. In this paper, we investigate the problem of identifying characters & annotating them with respective name using graph matching algorithm to get the most accurate identification result. The contribution of our work include: 1) the character-character relationship representation including a noise insensitive 2) Use an edit operation based error correcting graph matching algorithm. 3) Graph partitioning and graph matching for more complex character changes to handle simultaneously. 4) The existing character identification approaches, also we are going to perform an in-depth sensitivity analysis which will introduce two types of simulated noises.