Self-Supervised Depth Completion for Active Stereo

Depth maps are a way to visualize the distance of objects around a sensor and are essential to allow robots to build accurate maps of a space and plan their paths for safe navigation. Knowing whether an object is close, and precisely how close, is crucial to avoid collisions. It is something that humans and animals do naturally – from our earliest moments we are learning through trial and error to understand how far things are away from us. For robots, Supervised AI training achieves a similar result by showing millions of images all precisely labelled with ‘ground truth’ so that the AI learns to recognize objects at different distances. But this is expensive, and the datasets needed are often rare and insufficient. Our new depth completion algorithms leverage a specialist neural network which works in conjunction with our stereo vision SLAM system to greatly improve the accuracy and quality of depth maps. Read more in our blog: https://www.slamcore.com/blog/turbo-charging-depth-maps-with-ai and paper: https://arxiv.org/abs/2110.03234 About SLAMcore Robots, drones and autonomous machines need to efficiently locate themselves, map and understand their surroundings in real-time. SLAMcore provides commercial-grade Spatial Intelligence software, which is accurate, robust and computationally efficient to manufacturers of robots, consumer electronics and devices worldwide. Simultaneous localization and mapping (SLAM) remains one of the hardest challenges for robot/product developers, consuming significant time, resources and hardware. SLAMcore’s world-class visual inertial SLAM software, runs locally in real-time, on cost-effective hardware, delivering robust real-world performance in days instead of years. SLAMcore software works out-of-the box with standard sensors/hardware and can be ported to a wide range of custom sensors/compute, letting developers concentrate on developing great products rather than struggling to solve real-time mapping and localization challenges. https://www.slamcore.com