Object Hallucination in Image Captioning
@inproceedings{Rohrbach2018ObjectHI, title={Object Hallucination in Image Captioning}, author={Anna Rohrbach and Lisa Anne Hendricks and Kaylee Burns and Trevor Darrell and Kate Saenko}, booktitle={EMNLP}, year={2018} }
Despite continuously improving performance, contemporary image captioning models are prone to "hallucinating" objects that are not actually in a scene. [...] Key Result Our analysis yields several interesting findings, including that models which score best on standard sentence metrics do not always have lower hallucination and that models which hallucinate more tend to make errors driven by language priors.Expand Abstract
Figures, Tables, and Topics from this paper
Paper Mentions
50 Citations
More Grounded Image Captioning by Distilling Image-Text Matching Model
- Computer Science
- 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
- 2020
- 8
- PDF
Egoshots, an ego-vision life-logging dataset and semantic fidelity metric to evaluate diversity in image captioning models
- Computer Science
- ICLR 2020
- 2020
- 3
- PDF
Learning to Generate Grounded Image Captions without Localization Supervision
- Computer Science
- ArXiv
- 2019
- 4
- PDF
Learning to Generate Grounded Visual Captions Without Localization Supervision
- Computer Science
- ECCV
- 2020
- 6
- PDF
REO-Relevance, Extraness, Omission: A Fine-grained Evaluation for Image Captioning
- Computer Science
- EMNLP/IJCNLP
- 2019
- 2
- PDF
References
SHOWING 1-10 OF 24 REFERENCES
Learning to Evaluate Image Captioning
- Computer Science
- 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition
- 2018
- 45
- PDF
Deep Compositional Captioning: Describing Novel Object Categories without Paired Training Data
- Computer Science
- 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
- 2016
- 182
- PDF
Discriminability Objective for Training Descriptive Captions
- Computer Science
- 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition
- 2018
- 86
- Highly Influential
- PDF
Neural Baby Talk
- Computer Science
- 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition
- 2018
- 208
- Highly Influential
- PDF
Understanding Blind People's Experiences with Computer-Generated Captions of Social Media Images
- Computer Science
- CHI
- 2017
- 58
- PDF
Speaking the Same Language: Matching Machine to Human Captions by Adversarial Training
- Computer Science
- 2017 IEEE International Conference on Computer Vision (ICCV)
- 2017
- 72
- PDF
CIDEr: Consensus-based image description evaluation
- Computer Science
- 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
- 2015
- 1,448
- Highly Influential
- PDF
Visual Genome: Connecting Language and Vision Using Crowdsourced Dense Image Annotations
- Computer Science
- International Journal of Computer Vision
- 2016
- 1,641
- PDF