Okay, here's the deal:
Live view is the litmus test of autofocus. You'll need a tripod to properly test your camera, so get one if you don't have one.
Put the camera on the tripod and set up your test target. Then autofocus using the center point.
Now switch into live view and go to maximum magnification. Examine the sharpness of the center area at maximum magnification.
Now it's time to see if the autofocus is failing to focus properly. If your lens has full time manual focus, just turn the focus ring until the image is as sharply focused as you can make it. If it doesn't have full time manual, then it'll have a switch to set the lens to manual focus mode. Put the lens in manual focus mode and then turn the focus ring until the image is as sharp as you can get it.
If the image is significantly sharper, then either your camera or your lens is miscalibrated.
To tell the difference between a miscalibrated lens and a miscalibrated camera, perform the same test with different lenses. If they're all off, then it's almost certainly the camera.
You can even figure out whether it's frontfocusing or backfocusing using this method. To do this, Perform the steps as before, but after autofocusing and, if necessary, putting the lens in manual focus mode, slowly turn the focus ring in the direction towards infinity (most lenses have a distance indicator to make it possible to know which direction to turn the ring). If the image gets sharper as you turn the ring, even if only briefly (meaning, it hits optimal focus and then gets worse), then your lens+camera combo is frontfocusing (meaning: it is actually putting focus in front of the target). If it just gets worse, it means your combo is backfocusing (meaning: it's putting focus behind the target).
If turning the focus ring towards infinity just makes things worse (thus supporting the hypothesis that it's backfocusing), then you can verify that it's backfocusing by performing the autofocus step again, and then turning the focus ring slowly away from infinity (the opposite direction of the direction you turned it previously). The image should get better and then worse. If the image also just gets worse (meaning the previous test indicated backfocusing, and this test indicates frontfocusing), then it means that the autofocus is actually getting you close to optimum focus.
If the fuzzy image shown previously is the best you can get even after manually focusing in live view, then it means that there's definitely a problem with the camera.