Simple Image Description Generator via a Linear Phrase-Based Approach

TL;DR

This paper introduces a simple, bilinear, phrase-based model for image captioning that leverages CNN features and syntax-aware language modeling to generate relevant descriptions with competitive performance.

Contribution

The paper presents a novel, straightforward bilinear approach focusing on syntax, achieving comparable results to complex models on the COCO dataset.

Findings

Achieves competitive captioning results on COCO dataset

Uses a simple bilinear model with CNN features and syntax-based language modeling

Demonstrates effectiveness of a less complex approach for image description

Abstract

Generating a novel textual description of an image is an interesting problem that connects computer vision and natural language processing. In this paper, we present a simple model that is able to generate descriptive sentences given a sample image. This model has a strong focus on the syntax of the descriptions. We train a purely bilinear model that learns a metric between an image representation (generated from a previously trained Convolutional Neural Network) and phrases that are used to described them. The system is then able to infer phrases from a given image sample. Based on caption syntax statistics, we propose a simple language model that can produce relevant descriptions for a given test image using the phrases inferred. Our approach, which is considerably simpler than state-of-the-art models, achieves comparable results on the recently release Microsoft COCO dataset.