Implementation for "Multilevel Language and Vision Integration for Text-to-Clip Retrieval" - View it on GitHub