Dear Neha, what do you mean really by "human body frequency"?
If the human is awake, muscles mobilize against vibration inputs, it is thus to be considered as a closed loop controlled system and the notion of "natural frequencies" is inapplicable. Some local vibrations remain out of control, but then comes the inter-individual variations which are very large...
If you seek for a precise advice, please further develop your question! (what sort of vibration exposure, which population is exposed, what sort of body response you intend to factor, what is your modelling aim, what sort of context - posture, environment, loads, protection equipment...)
A human can stand, sit or lay on a suiteable underlay mounted on a shaker capable of driving a mass like the human body, and sensors can be monted on different parts of the body. Transfer functions between the shaker force and recorded acceleration, or velocity, will reveal the resonances of the different body parts. They may differ from body to body. And a test would probaly need to be in all three directions.