The recognition of individuals through their walking pattern, known as Human Gait Recognition, is a popular biometric approach. The way humans walk can be affected by factors such as angle variation, clothing variation, foot shadows, and carrying conditions. Enhancing frontal gait recognition performance is the main challenge in the appearance-based gait approach. The absence of spatio-temporal information sets this task apart from the other view variations, making it more challenging. The scope of this study encompasses a successful approach to identifying individuals in frontal-view gait sequences. By employing frontal Gait Energy Image (GEI), the proposed method acquires efficient feature vectors at the beginning. Next, the extracted features go through several regressors before being combined and reduced using a score-based feature reduction method. Improved gait recognition performance is achieved through the spatial dynamics captured by the reduced feature’s ability. The effectiveness of the proposed approach was assessed using well-known gait datasets like CMU MoBo, CASIA A, and CASIA B. The experiments demonstrated that the proposed approach yielded promising results and outperformed some state-of-the-art approaches in recognition.