Numerous published studies have confirmed the potentials of deploying gait as biometrics within forensic and surveillance scenarios. Few of these have addressed the contribution of motion-based features on the recognition process. We describe in this paper a descriptor based on computing the optical flow of consecutive frames to generative a discriminative biometric signature for gait recognition. A set of experiments are carried out using the CASIA dataset to explore the usefulness of motion-based features for gait identification subjected to different covariate factors including clothing and carrying conditions. Based on a limited dataset containing 100 samples, higher recognition rates are achieved using plain motion features using the nearest neighbor classifier. The attained results affirm that people identification via the use of features from gait kinematics is achievable with acceptable success rates regardless of the covariate factors. This is a crucial milestone in shifting academic studies on gait biometrics from research settings to forensic and surveillance environments.