Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

CLIPScore: A Reference-free Evaluation Metric for Image Captioning

About

Image captioning has conventionally relied on reference-based automatic evaluations, where machine captions are compared against captions written by humans. This is in contrast to the reference-free manner in which humans assess caption quality. In this paper, we report the surprising empirical finding that CLIP (Radford et al., 2021), a cross-modal model pretrained on 400M image+caption pairs from the web, can be used for robust automatic evaluation of image captioning without the need for references. Experiments spanning several corpora demonstrate that our new reference-free metric, CLIPScore, achieves the highest correlation with human judgements, outperforming existing reference-based metrics like CIDEr and SPICE. Information gain experiments demonstrate that CLIPScore, with its tight focus on image-text compatibility, is complementary to existing reference-based metrics that emphasize text-text similarities. Thus, we also present a reference-augmented version, RefCLIPScore, which achieves even higher correlation. Beyond literal description tasks, several case studies reveal domains where CLIPScore performs well (clip-art images, alt-text rating), but also where it is relatively weaker in comparison to reference-based metrics, e.g., news captions that require richer contextual knowledge.

Jack Hessel, Ari Holtzman, Maxwell Forbes, Ronan Le Bras, Yejin Choi• 2021

Related benchmarks

TaskDatasetResultRank
Image ClassificationImageNet 1k (test)
Top-1 Accuracy27.3
798
Image Captioning EvaluationComposite
Kendall-c Tau_c57.3
92
Image Captioning EvaluationFlickr8K Expert (test)
Kendall tau_c53
76
Image Captioning EvaluationFlickr8k Expert
Kendall Tau-c (tau_c)54.3
73
Image Captioning EvaluationPascal-50S (test)
HC64.9
66
Perceptual Quality AssessmentHPE-Bench 1.0 (test)
SRCC0.2446
66
Image Captioning EvaluationFlickr8K-CF (test)
Kendall tau_b36.4
65
Image Captioning EvaluationFlickr8K-CF
Kendall-b Correlation (tau_b)36.6
62
Image Captioning EvaluationPascal-50S
Mean Score84.6
39
Editing Alignment AssessmentHPE-Bench 1.0 (test)
SRCC0.2099
33
Showing 10 of 69 rows

Other info

Follow for update