Learning to Evaluate Image Captioning

Abstract

Evaluation metrics for image captioning face two challenges. Firstly, commonly used metrics such as CIDEr, METEOR, ROUGE and BLEU often do not correlate well with human judgments. Secondly, each metric has well known blind spots to pathological caption constructions, and rule-based metrics lack provisions to repair such blind spots once identified. For example, the newly proposed SPICE correlates well with human judgments, but fails to capture the syntactic structure of a sentence. To address these two challenges, we propose a novel learning based discriminative evaluation metric that is directly trained to distinguish between human and machine-generated captions. In addition, we further propose a data augmentation scheme to explicitly incorporate pathological transformations as negative examples during training. The proposed metric is evaluated with three kinds of robustness tests and its correlation with human judgments. Extensive experiments show that the proposed data augmentation scheme not only makes our metric more robust toward several pathological transformations, but also improves its correlation with human judgments. Our metric outperforms other metrics on both caption level human correlation in Flickr 8k and system level human correlation in COCO. The proposed approach could be served as a learning based evaluation metric that is complementary to existing rule-based metrics.

CVPR 2018

References in corpus (13)

• Show, Attend and Tell: Neural Image Caption Generation with Visual Attention
• Deep Residual Learning for Image Recognition
• Microsoft COCO Captions: Data Collection and Evaluation Server
• Improved Image Captioning via Policy Gradient optimization of SPIDEr
• Multimodal Compact Bilinear Pooling for Visual Question Answering and Visual Grounding
• Adversarial Learning for Neural Dialogue Generation
• Show and Tell: A Neural Image Caption Generator
• Towards Diverse and Natural Image Descriptions via a Conditional GAN
• Recurrent Topic-Transition GAN for Visual Paragraph Generation
• Adversarial Evaluation of Dialogue Models
• CIDEr: Consensus-based Image Description Evaluation
• Compact Bilinear Pooling
• Speaking the Same Language: Matching Machine to Human Captions by Adversarial Training

Cited by in corpus (5)

• Evaluation of Text Generation: A Survey
• Gated Hierarchical Attention for Image Captioning
• Dynamic population-based meta-learning for multi-agent communication with natural language
• Intrinsic Image Captioning Evaluation
• Reason induced visual attention for explainable autonomous driving