Publication | Open Access

MKEAH： Multimodal knowledge extraction and accumulation based on hyperplane embedding for knowledge-based visual question answering