Computer Vision News - June 2020

3 Summary 3D Poses in the Wild 9 build an intelligence system that can reason about the world and think like humans do,” Gerard tells us. “We can’t think without our body. My son is only five months old and I’m seeing that developed yet, but the kinds of things he can do are constrained by his body. He can’t grasp every object, he can’t walk, and I think for adults, it’s the same. The way we think is linked to the body that is carrying our intelligence system. Estimating the 3D pose of people is just one aspect of this. That’s why I find this so exciting.” Angjoo agrees: “We live in a 3D world. It’s actually 4D because we are moving over time and we interact with other people. That’s what we really want our intelligence systems to do. If they want to interact and understand what other people are doing, the first step is to be able to see where they are and what they are doing in 3D. Understanding that is a first step, I think. That’s why it’s very exciting to me.” With numerous possible applications of this work, there are a wide range of groups who are waiting for good solutions to human pose estimation. From medicine, biomechanics, and rehabilitation, to entertainment, graphics, video gaming, and virtual reality. The list goes on. Like us, they will all be very keen to see the results. Curious to learn about our story? Find it out in the Research of the Month at page 12