Ozbalkan, UgurTurna, Ozgur Can2026-02-102026-02-1020262076-341710.3390/app160104882-s2.0-105027294074https://doi.org/10.3390/app16010488https://hdl.handle.net/20.500.14627/1408This study investigates the performance of three deep learning architectures-LSTM with Attention, GRU with Attention, and Transformer-in the context of real-time, self-guided exercise classification, using coordinate data collected from 103 participants via a dual-camera system. Each model was evaluated over ten randomized runs to ensure robustness and statistical validity. The GRU + Attention and LSTM + Attention models demonstrated consistently high test accuracy (mean approximate to 98.9%), while the Transformer model yielded significantly lower accuracy (mean approximate to 96.6%) with greater variance. Paired t-tests confirmed that the difference between LSTM and GRU models was not statistically significant (p = 0.9249), while both models significantly outperformed the Transformer architecture (p < 0.01). In addition, participant-specific features, such as athletic experience and BMI, were found to affect classification accuracy. These findings support the feasibility of AI-based feedback systems in enhancing unsupervised training, offering a scalable solution to bridge the gap between expert supervision and autonomous physical practice.eninfo:eu-repo/semantics/openAccessArtificial Intelligence in Exercise MonitoringDeep Learning for Exercise FeedbackHuman Motion TrackingReal-Time Movement ClassificationSelf-Guided Fitness TrainingArticle