How Good Is Google Bard’s Visual Understanding? An Empirical Study On Open Challenges

This study, published in Machine Intelligence Research, evaluates Google Bard, a generative AI chatbot that accepts prompts and performs text-based tasks such as providing answers, summaries, and creating various forms of text content. The study focuses on Bard’s capability to analyze visual content and provide a description or answer questions using visual information. The researchers’ goal is to analyze the capability of Bard towards some of the long-standing problems of computer vision in image comprehension. The evaluation does not include quantitative results on large-scale benchmarks, but rather focuses on identifying a number of insightful scenarios and corresponding visual-textual prompts to evaluate not only the visual understanding capabilities of Bard but future large multimodal models such as GPT4 as well.