Some yes, but mostly no.
Yes: A class itself is not an object. It's the structure for an object. For it to become an object, it needs to be instantiated with the "new" function:
$stefan = new person();
No: Basically everything else. Whether you are accessing an object's properties directly (example: "$stefan->name;"), or using a getter method (example: "$stefan->get_name()"), either way, you are accessing an object that has been created.
It is technically possible to access properties/methods without instantiating them in an object using the "static" keyword (http://php.net/manual/en/language.oop5.static.php), but that isn't really within the scope of your comment above.
---
Basically, the point is this. Say you create an object, and it has a property that holds a value. The object itself should be in control of the value of all of its properties, and it should have the option of rejecting outside changes if necessary. You shouldn't make it possible for outside code to change the value of a property without the object knowing about it. This is especially important if the object is expecting a property to hold a specific value -- if an outside source changes it, it may lead to bugs in your code due to incorrect expectations on the value of the property.
By using setters/getters, you are keeping the control within the object itself. For example, say in a setter method, you could set up restrictions so that certain values could not be passed in. For example, take the Person class that was created in the example on killerphp.com. I think we can safely assume that if someone is a person, they have a name. So, in the setter method for the object, you could include code that would reject any attempt to clear the person's name:
function set_name($new_name) {
if ($new_name != '') {
$this->name = $new_name;
}
}
This way, the object itself has control over its properties, and it's properties can't be unexpectedly changed, or changed in a way that wouldn't make sense. Hopefully that makes sense?