Stacked Cross-modal Feature Consolidation Attention Networks for Image Captioning

TL;DR

This paper introduces a novel stacked cross-modal feature consolidation attention network for image captioning, integrating high-level semantics and visual cues through multi-step reasoning to generate more detailed captions.

Contribution

It proposes a new SCFC attention network with a compounding function and context-aware attributes, advancing end-to-end image captioning methods.

Findings

Outperforms state-of-the-art on MSCOCO and Flickr30K datasets

Effectively combines semantic and visual information for detailed captions

Uses a novel multi-step reasoning approach for feature consolidation

Abstract

Recently, the attention-enriched encoder-decoder framework has aroused great interest in image captioning due to its overwhelming progress. Many visual attention models directly leverage meaningful regions to generate image descriptions. However, seeking a direct transition from visual space to text is not enough to generate fine-grained captions. This paper exploits a feature-compounding approach to bring together high-level semantic concepts and visual information regarding the contextual environment fully end-to-end. Thus, we propose a stacked cross-modal feature consolidation (SCFC) attention network for image captioning in which we simultaneously consolidate cross-modal features through a novel compounding function in a multi-step reasoning fashion. Besides, we jointly employ spatial information and context-aware attributes (CAA) as the principal components in our proposed compounding function, where our CAA provides a concise context-sensitive semantic representation. To make better use of consolidated features potential, we further propose an SCFC-LSTM as the caption generator, which can leverage discriminative semantic information through the caption generation process. The experimental results indicate that our proposed SCFC can outperform various state-of-the-art image captioning benchmarks in terms of popular metrics on the MSCOCO and Flickr30K datasets.