At the request of the authors, this paper has been retracted. The authors identified significant errors in the methodology and figures presented in the article. Specifically, they discovered that the facial extraction method described in the paper was incorrectly attributed to dlib, whereas the method used was MTCNN. The misrepresentation affects the validity and reproducibility of the results presented in the study. An inaccuracy in Fig. 2. which illustrates the “Key Frame Extraction” process, depicts the process as being based on the residual relative to the first frame, when in fact, it should be based on the residual relative to the average image of all key frame sequences. The errors fundamentally affect the integrity of the research and its findings. The misrepresentation of the methods and the incorrect figure misled readers and the scientific community regarding the novelty and effectiveness of the methods the authors claimed to have developed. In light of the omissions and accuracies, the authors believe it is their ethical responsibility to retract the paper.