ICLR 2023 - F-VLM: Open-Vocabulary Object Detection upon Frozen Vision and Language Models | AI Breakdown | Podwise