vidas = sees. vidas = watches?
vidas = sees, rigardas = watches.
Rigardas can also mean "looks at", right?
Yes, that's true.
Why is it bebon and not bebo?
Because it's the direct object of the verb vidas.
Like how in English I would say "I see him" and not "I see he" -- the thing seen is the direct object.
In Esperanto, the direct object (generally) gets a -n for the accusative case.
Hence, La knabo vidas la bebon.
Isn't "etulo" for baby more common or at least also acceptable?
I think "etulo" is used specifically for a toddler. "bebo" refers to an infant, or "baby".
Wiktionary