Beyond RNNs: Benchmarking Attention-Based Image Captioning Models

TL;DR

This paper benchmarks attention-based image captioning models against RNN-based approaches, demonstrating that attention mechanisms significantly improve caption quality and alignment with human judgments on the MS-COCO dataset.

Contribution

It provides a comprehensive comparison of attention-based and RNN-based models for image captioning, highlighting the effectiveness of attention mechanisms in improving caption accuracy and semantic richness.

Findings

Attention models outperform RNNs in caption quality.

Attention improves alignment with human evaluations.

Models achieve higher scores on BLEU, METEOR, GLEU, and WER metrics.

Abstract

Image captioning is a challenging task at the intersection of computer vision and natural language processing, requiring models to generate meaningful textual descriptions of images. Traditional approaches rely on recurrent neural networks (RNNs), but recent advancements in attention mechanisms have demonstrated significant improvements. This study benchmarks the performance of attention-based image captioning models against RNN-based approaches using the MS-COCO dataset. We evaluate the effectiveness of Bahdanau attention in enhancing the alignment between image features and generated captions. The models are assessed using natural language processing metrics such as BLEU, METEOR, GLEU, and WER. Our results show that attention-based models outperform RNNs in generating more accurate and semantically rich captions, with better alignment to human evaluation. This work provides insights into the impact of attention mechanisms in image captioning and highlights areas for future improvements.