Let there be a clock on the beach: Reducing Object Hallucination in Image Captioning
This addresses a common problem in state-of-the-art captioning models that is undesirable for human users, though it appears incremental as it builds on existing methods.
The paper tackles object hallucination in image captioning, where models describe missing or non-existent objects, by proposing three training augmentation methods that reduce this bias without requiring new data or larger models, achieving significant improvements on hallucination metrics.
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 increase in 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 will be made public.