This study deals with the concept of Uncanny Valley identified by the Japanese robotics professor, Masahiro Mori. Following this concept introduced by Mori for robots, three-dimensional computer-generated images that are produced in computer environment and used in animation films are analyzed. In order to overcome the line that causes difference between live and computer-generated images, three-dimensional characters having similar features to live characters are created. While some of the animation films produced in accordance with this aim are considered successful, the others are not able to achieve desired goals and they are not admired by the audience. In this respect, it is highly important whether film characters are found interesting or repulsive. It is also significant whether those films fall into the category of Uncanny Valley or not. In accordance with this purpose, physical appearance and movements of three dimensional characters in the animation films, Beowulf (2007, Robert Zemeckis) and Frozen (2013, Chris Buck, Jennifer Lee) have been examined by using focus groups so as to detect whether these two films fall into Uncanny Valley or not. In the study by evaluating data obtained through focus groups, the movements and physical appearance of characters of the film Beowulf that cause the film to fall into the category of Uncanny Valley have been analyzed. The results of the study show that the eyes of the characters in the movies took the lead in falling into Uncanny Valley. The reason behind this situation can be explained by the great impact of the frozen expression in the eyes of the characters. On the other hand, the kind of movies like Frozen cannot be considered within the concept of Uncanny Valley because of their linear structure.