Comparing Automatic Evaluation Measures for Image Description
Open accessThe DOI matches its Crossref record; 3 of the 3 references it cites were found in a registry or an index.
Abstract
Image description is a new natural language generation task, where the aim is to generate a human-like description of an image.The evaluation of computer-generated text is a notoriously difficult problem, however, the quality of image descriptions has typically been measured using unigram BLEU and human judgements.The focus of this paper is to determine the correlation of automatic measures with human judgements for this task.We estimate the correlation of unigram and Smoothed BLEU, TER, ROUGE-SU4, and Meteor against human judgements on two data sets.The main finding is that unigram BLEU has a weak correlation, and Meteor has the strongest correlation with human judgements.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
BLEU: a method for automatic evaluation of machine translation
Papineni, K., Roukos, S., Ward, T., & Zhu, W.-J. (2001) · Proceedings of the 40th Annual Meeting on Association for Computational Linguistics - ACL '02 · doi:10.3115/1073083.1073135 · Cited by 21,344
Object Detection with Discriminatively Trained Part-Based Models
Felzenszwalb, P. F., Girshick, R. B., McAllester, D., & Ramanan, D. (2010) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2009.167 · Cited by 9,853
Lin, C.-Y., & Och, F. J. (2004) · Proceedings of the 42nd Annual Meeting on Association for Computational Linguistics - ACL '04 · doi:10.3115/1218955.1219032 · Cited by 737 · cited 1×
“ROUGE measures the longest common subsequence of tokens between a candidate Y and reference X (Lin and Och, 2004).”
Background
Cited by
Multimodal Machine Learning: A Survey and Taxonomy
Baltrusaitis, T., Ahuja, C., & Morency, L.-P. (2019) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2018.2798607 · Cited by 5,018
CIDEr: Consensus-based image description evaluation
Vedantam, R., Zitnick, C. L., & Parikh, D. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7299087 · Cited by 4,845
VQA: Visual Question Answering
Antol, S., Agrawal, A., Lu, J., Mitchell, M., Batra, D., Zitnick, C. L., & Parikh, D. (2015) · 2015 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2015.279 · Cited by 4,487
SPICE: Semantic Propositional Image Caption Evaluation
Anderson, P., Fernando, B., Johnson, M., & Gould, S. (2016) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-46454-1_24 · Cited by 1,986
Microsoft COCO Captions: Data Collection and Evaluation Server
Chen, X., Fang, H., Lin, T.-Y., Vedantam, R., Gupta, S., Dollar, P., & Zitnick, C. L. (2015) · arXiv · doi:10.48550/arxiv.1504.00325 · Cited by 1,614
“Designing effective automatic evaluation metrics that are highly correlated with human judgment remains a difficult challenge [7], [42], [45], [46].”
BackgroundFraming Image Description as a Ranking Task: Data, Models and Evaluation Metrics
Hodosh, M., Young, P., & Hockenmaier, J. (2013) · Journal of Artificial Intelligence Research · doi:10.1613/jair.3994 · Cited by 1,314
From captions to visual concepts and back
Fang, H., Gupta, S., Iandola, F., Srivastava, R. K., Deng, L., Dollar, P., Gao, J., He, X., Mitchell, M., Platt, J. C., et al. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7298754 · Cited by 1,287
CLIPScore: A Reference-free Evaluation Metric for Image Captioning
Hessel, J., Holtzman, A., Forbes, M., Le Bras, R., & Choi, Y. (2021) · Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/2021.emnlp-main.595 · Cited by 1,232
Translating Videos to Natural Language Using Deep Recurrent Neural Networks
Venugopalan, S., Xu, H., Donahue, J., Rohrbach, M., Mooney, R., & Saenko, K. (2015) · Proceedings of the 2015 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies · doi:10.3115/v1/n15-1173 · Cited by 959
VQA: Visual Question Answering
Agrawal, A., Lu, J., Antol, S., Mitchell, M., Zitnick, C. L., Batra, D., & Parikh, D. (2015) · arXiv (Cornell University) · doi:10.48550/arxiv.1505.00468 · Cited by 900
Gatt, A., & Krahmer, E. (2018) · Journal of Artificial Intelligence Research · doi:10.1613/jair.5477 · Cited by 776
“As for the relationship between metrics and human judgements, Elliott and Keller (2013) concluded that meteor correlates better than bleu (see Elliott & Keller, 2014, for a systematic comparison of automatic metrics in this domain), a finding also confirmed in their later work (Elliott & De Vries,…”
BackgroundResultVQA: Visual Question Answering: www.visualqa.org
Agrawal, A., Lu, J., Antol, S., Mitchell, M., Zitnick, C. L., Parikh, D., & Batra, D. (2017) · International Journal of Computer Vision · doi:10.1007/s11263-016-0966-6 · Cited by 570
Mind's eye: A recurrent visual representation for image caption generation
Chen, X., & Zitnick, C. L. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7298856 · Cited by 498
Why We Need New Evaluation Metrics for NLG
Novikova, J., Dušek, O., Cercas Curry, A., & Rieser, V. (2017) · Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/d17-1238 · Cited by 422
Why We Need New Evaluation Metrics for NLG
Novikova, J., Dušek, O., Curry, A. C., & Rieser, V. (2017) · arXiv · doi:10.48550/arxiv.1707.06875 · Cited by 406
Deep correlation for matching images and text
Yan, F., & Mikolajczyk, K. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7298966 · Cited by 395
Automatic Description Generation from Images: A Survey of Models, Datasets, and Evaluation Measures
Bernardi, R., Cakici, R., Elliott, D., Erdem, A., Erdem, E., Ikizler-Cinbis, N., Keller, F., Muscat, A., & Plank, B. (2016) · Journal of Artificial Intelligence Research · doi:10.1613/jair.4900 · Cited by 349
Self-citationT ree T alk: Composition and Compression of Trees for Image Descriptions
Kuznetsova, P., Ordonez, V., Berg, T. L., & Choi, Y. (2014) · Transactions of the Association for Computational Linguistics · doi:10.1162/tacl_a_00188 · Cited by 267
Situation Recognition: Visual Semantic Role Labeling for Image Understanding
Yatskar, M., Zettlemoyer, L., & Farhadi, A. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.597 · Cited by 242
“However, we sidestep the evaluation challenges that come with generating sentences [45, 7], while also providing visual evidence for verbs that should aid captioning.”
A Shared Task on Multimodal Machine Translation and Crosslingual Image Description
Specia, L., Frank, S., Sima'an, K., & Elliott, D. (2016) · Proceedings of the First Conference on Machine Translation: Volume 2, Shared Task Papers · doi:10.18653/v1/w16-2346 · Cited by 227
“Meteor is the primary evaluation measure because it has been shown to have a much stronger correlation with human judgements than BLEU or TER for this task (Elliott and Keller, 2014).”
BackgroundResultSelf-citationA Survey of Evaluation Metrics Used for NLG Systems
Sai, A. B., Mohankumar, A. K., & Khapra, M. M. (2023) · ACM Computing Surveys · doi:10.1145/3485766 · Cited by 235
Good News, Everyone! Context Driven Entity-Aware Captioning for News Images
Biten, A. F., Gomez, L., Rusiñol, M., & Karatzas, D. (2019) · 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2019.01275 · Cited by 136
“As discussed extensively in the literature [8, 10, 17, 38, 6] standard evaluation metrics for image captioning have several flaws and in many cases they do not correlate with human judgments.”
BackgroundGan, C., Li, Y., Li, H., Sun, C., & Gong, B. (2017) · 2017 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.201 · Cited by 116
“…in the image that can visually answer the question. jective — it is hard to evaluate the quality of captions generated by different algorithms [7, 41, 1], and tend to miss subtle details — in training, the models may be led to capturing the scene-level gist rather than fine-grained entities.”
BackgroundTIGEr: Text-to-Image Grounding for Image Caption Evaluation
Jiang, M., Huang, Q., Zhang, L., Wang, X., Zhang, P., Gan, Z., Diesner, J., & Gao, J. (2019) · Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP) · doi:10.18653/v1/d19-1220 · Cited by 59
Learning to Ask Like a Physician
Lehman, E., Lialin, V., Legaspi, K. E., Sy, A. J., Pile, P. T., Alberto, N. R., Ragasa, R. R., Puyat, C. V., Taliño, M. K., Alberto, I. R., et al. (2022) · Proceedings of the 4th Clinical Natural Language Processing Workshop · doi:10.18653/v1/2022.clinicalnlp-1.8 · Cited by 10
Crowdsourcing Thumbnail Captions: Data Collection and Validation
Aguirre, C., Cao, S., Mahmood, A., & Huang, C.-M. (2023) · ACM Transactions on Interactive Intelligent Systems · doi:10.1145/3589346 · Cited by 6
HallE-Control: Controlling Object Hallucination in Large Multimodal Models
Zhai, B., Yang, S., Xu, C., Shen, S., Keutzer, K., Li, C., & Li, M. (2023) · arXiv · doi:10.48550/arxiv.2310.01779 · Cited by 2
Metadata
- Authors
- Elliott, D., & Keller, F.
- Venue
- Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers)
- Published
- 2014
- Pages
- 452-457
- OpenAlex
- W2128856065
- Cited by
- 134
- References
- 3
- FWCI
- 11.46
- Type
- conference paper
- Language
- en
- Open access
- hybrid
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.