Today, we have an interview with Naila Murray, director of AI research at Meta, to launch our AI Trends 2024 series. We delve into the most recent advancements and trends in the field of computer vision in our chat with Naila. We discuss developments in multimodal models (vision plus LLMs), 3D Gaussian splatting, visual programming, and controlled generation. We talk about tools and open-source projects such as DINOv2, a visual encoding model that allows object recognition, segmentation, and depth estimation even in data-scarce scenarios, ControlNet, which adds conditional control to stable diffusion models, and Segment Anything, a tool for versatile zero-shot image segmentation using simple text prompts clicks, and bounding boxes. Lastly, Naila discusses her thoughts on the most promising prospects in the industry and her forecasts for the next years.
