Text Data-Centric Image Captioning with Interactive Prompts

arXiv:2403.19193v1 Announce Type: new
Abstract: Supervised image captioning approaches have made great progress, but it is challenging to collect high-quality human-annotated image-text data. Recently, large-scale vision and language models (e.g., CLIP) and large-scale generative language models (e.g., GPT-2) have shown strong performances in various tasks, which also provide some new solutions for image captioning with web paired data, unpaired data or even text-only data. Among them, the mainstream solution is to project image embeddings into the text embedding space with …

abstract arxiv captioning clip cs.cv data data-centric generative gpt gpt-2 human image interactive language language models performances progress prompts quality scale solutions tasks text type vision