PRISM: Probabilistic Real-Time Inference in Spatial World Models
This work addresses the need for real-time, probabilistic SLAM with uncertainty estimates for applications like UAVs and handheld cameras, though it is incremental as it builds on existing methods.
The authors tackled the problem of real-time probabilistic inference for simultaneous localization and mapping (SLAM) by introducing PRISM, which reconciles uncertainty estimates, real-time operation, dense scene representation, and agent dynamics modeling, achieving 10Hz real-time performance with accuracy comparable to state-of-the-art SLAM on datasets like Blackbird, EuRoC, and TUM-RGBD.
We introduce PRISM, a method for real-time filtering in a probabilistic generative model of agent motion and visual perception. Previous approaches either lack uncertainty estimates for the map and agent state, do not run in real-time, do not have a dense scene representation or do not model agent dynamics. Our solution reconciles all of these aspects. We start from a predefined state-space model which combines differentiable rendering and 6-DoF dynamics. Probabilistic inference in this model amounts to simultaneous localisation and mapping (SLAM) and is intractable. We use a series of approximations to Bayesian inference to arrive at probabilistic map and state estimates. We take advantage of well-established methods and closed-form updates, preserving accuracy and enabling real-time capability. The proposed solution runs at 10Hz real-time and is similarly accurate to state-of-the-art SLAM in small to medium-sized indoor environments, with high-speed UAV and handheld camera agents (Blackbird, EuRoC and TUM-RGBD).