TY  - CONF
AU  - Ali Furkan Biten
AU  - Lluis Gomez
AU  - Dimosthenis Karatzas
A2  - WACV
PY  - 2022//
TI  - Let there be a clock on the beach: Reducing Object Hallucination in Image Captioning
BT  - Winter Conference on Applications of Computer Vision
SP  - 1381
EP  - 1390
KW  - Measurement
KW  - Training
KW  - Visualization
KW  - Analytical models
KW  - Computer vision
KW  - Computational modeling
KW  - Training data
N2  - Explaining an image with missing or non-existent objects is known as object bias (hallucination) in image captioning. This behaviour is quite common in the state-of-the-art captioning models which is not desirable by humans. To decrease the object hallucination in captioning, we propose three simple yet efficient training augmentation method for sentences which requires no new training data or increasein the model size. By extensive analysis, we show that the proposed methods can significantly diminish our models’ object bias on hallucination metrics. Moreover, we experimentally demonstrate that our methods decrease the dependency on the visual features. All of our code, configuration files and model weights are available online.
UR  - https://ieeexplore.ieee.org/document/9706727
L1  - http://refbase.cvc.uab.es/files/BGK2022.pdf
UR  - http://dx.doi.org/10.1109/WACV51458.2022.00253
N1  - DAG; 600.155; 302.105
ID  - Ali Furkan Biten2022
ER  -