Co-Fusion: Real-time Segmentation, Tracking and Fusion of Multiple Objects
This work addresses the challenge of dynamic scene understanding for robotics, allowing interactions in changing environments, though it builds incrementally on existing SLAM and segmentation techniques.
The paper tackles the problem of real-time 3D reconstruction in dynamic scenes by introducing Co-Fusion, a dense SLAM system that segments, tracks, and fuses multiple objects from RGB-D input, enabling robots to maintain object-level scene descriptions with improved accuracy and robustness.
In this paper we introduce Co-Fusion, a dense SLAM system that takes a live stream of RGB-D images as input and segments the scene into different objects (using either motion or semantic cues) while simultaneously tracking and reconstructing their 3D shape in real time. We use a multiple model fitting approach where each object can move independently from the background and still be effectively tracked and its shape fused over time using only the information from pixels associated with that object label. Previous attempts to deal with dynamic scenes have typically considered moving regions as outliers, and consequently do not model their shape or track their motion over time. In contrast, we enable the robot to maintain 3D models for each of the segmented objects and to improve them over time through fusion. As a result, our system can enable a robot to maintain a scene description at the object level which has the potential to allow interactions with its working environment; even in the case of dynamic scenes.