Video games have been around for a long time and have become a big part of people’s lives. The number of people that enjoy them are numerous; however, not everyone who enjoys them can actually play them due to physical limitations. This study aims to help people with disabilities, specifically those on the arms or hands, by providing them a new way to play games, specifically computer games, through the use of face detection, face recognition, and a face and head movement tracking system to control the video game. The program is made to be utilized in playing only simple games with at most seven controls, and can only input at maximum, two controls simultaneously. Special focus was given in making the program lightweight in order to be capable of real-time gameplay. The respondents for this study were sampled using the purposive sampling technique. The program was used by the respondents using computer vision and rated the program in accordance with ISO 25010. Their answers were obtained by utilizing surveys. On the other hand, the accuracy of the program was evaluated through alpha testing results utilizing the confusion matrix and real-world testing results. The results showed that the respondents strongly agree with the functional suitability, compatibility, and portability of the software, while they agree with the performance efficiency, usability, reliability, security, and maintainability of the software. Furthermore, the program yielded an average accuracy of 88% and an average F1 score of 0.86 or 86% in the alpha testing results, and an average accuracy of 89% for real world testing results. These results show that the program produced satisfactory results and can be used to play computer games hands-free.