EVCap: Retrieval-Augmented Image Captioning with External Visual-Name Memory for Open-World Comprehension

arXiv:2311.15879v2 Announce Type: replace
Abstract: Large language models (LLMs)-based image captioning has the capability of describing objects not explicitly observed in training data; yet novel objects occur frequently, necessitating the requirement of sustaining up-to-date object knowledge for open-world comprehension. Instead of relying on large amounts of data and/or scaling up network parameters, we introduce a highly effective retrieval-augmented image captioning method that prompts LLMs with object names retrieved from External Visual--name memory (EVCap). We build ever-changing object knowledge memory using …

abstract arxiv capability captioning cs.cv data image knowledge language language models large language large language models llms memory novel object objects open-world retrieval retrieval-augmented scaling training training data type visual world