Cross-Modal Food Retrieval: Learning a Joint Embedding of Food Images and Recipes With Semantic Consistency and Attention Mechanism | IEEE Journals & Magazine | IEEE Xplore