VSE++: Improving Visual-Semantic Embeddings with Hard Negatives

18 Jul 2017Fartash FaghriDavid J. FleetJamie Ryan KirosSanja Fidler

We present a new technique for learning visual-semantic embeddings for cross-modal retrieval. Inspired by hard negative mining, the use of hard negatives in structured prediction, and ranking loss functions, we introduce a simple change to common loss functions used for multi-modal embeddings... (read more)

PDF Abstract

Evaluation results from the paper


  Submit results from this paper to get state-of-the-art GitHub badges and help community compare results to other papers.