Should I use hyphenation on compounds consisting of three nouns?
Solution 1:
The hyphen is not needed because these are noun phrases and they are being used as nouns. They are not being turned into adjectival modifiers. You are detecting hand gestures, ergo hand gesture detection. You are extracting face features (though I'd recommend calling that facial features), ergo facial feature extraction. Just like with heavy metal detection, no hyphen is needed.
Compare with: He's a hand-gesture junkie. Wherein hand-gesture is modifying and describing what type of junkie he is, and acts as a single adjective.