Taken from the website:
"Instead of referring to objects (pics, articles, videos) by which server they are stored on, IPFS refers to everything by the hash on the file. The idea is that if in your browser you want to access a particular page then IPFS will ask the entire network “does anyone have this file that corresponds to this hash?” and a node on IPFS that does can return the file allowing you to access it."
I think this explains it best for noobs like myself :P
Yes, basically like a torrent network where the file your are downloading is replaced with... well, everything.
Every computer can be node on this network.
Clever system.