Deep Visual-Semantic Alignments for Generating Image Descriptions

Abstract

We present a model that generates natural language descriptions of images and their regions. Our approach leverages datasets of images and their sentence descriptions to learn about the inter-modal correspondences between language and visual data. Our alignment model is based on a novel combination of Convolutional Neural Networks over image regions, bidirectional Recurrent Neural Networks over sentences, and a structured objective that aligns the two modalities through a multimodal embedding. We then describe a Multimodal Recurrent Neural Network architecture that uses the inferred alignments to learn to generate novel descriptions of image regions. We demonstrate that our alignment model produces state of the art results in retrieval experiments on Flickr8K, Flickr30K and MSCOCO datasets. We then show that the generated descriptions significantly outperform retrieval baselines on both full images and on a new dataset of region-level annotations.

References in corpus (10)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Recurrent Neural Network Regularization
• Microsoft COCO Captions: Data Collection and Evaluation Server
• Going Deeper with Convolutions
• Unifying Visual-Semantic Embeddings with Multimodal Neural Language Models
• Explain Images with Multimodal Recurrent Neural Networks
• Learning a Recurrent Visual Representation for Image Caption Generation
• A Joint Model of Language and Perception for Grounded Attribute Learning
• CIDEr: Consensus-based Image Description Evaluation
• Video In Sentences Out

Cited by in corpus (16)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Fully Connected Deep Structured Networks
• Fisher Vectors Derived from Hybrid Gaussian-Laplacian Mixture Models for Image Annotation
• Recurrent Topic-Transition GAN for Visual Paragraph Generation
• Person Search with Natural Language Description
• A Dataset for Movie Description
• Visual Causal Feature Learning
• End-to-end people detection in crowded scenes
• Simple Image Description Generator via a Linear Phrase-Based Approach
• GuessWhat?! Visual object discovery through multi-modal dialogue
• Semantic Compositional Networks for Visual Captioning
• Generating Multi-Sentence Lingual Descriptions of Indoor Scenes
• Learning language through pictures
• Commonly Uncommon: Semantic Sparsity in Situation Recognition
• Hard to Cheat: A Turing Test based on Answering Questions about Images
• DeepDiary: Automatic Caption Generation for Lifelogging Image Streams