SCFusion: Real-time Incremental Scene Reconstruction with Semantic Completion
This addresses the issue of partial reconstructions for real-time applications such as augmented reality and robotics, though it appears incremental as it builds on existing methods by adding semantic completion.
The paper tackles the problem of incomplete 3D scene reconstruction from depth data due to occlusion, which limits real-time applications like augmented reality, by proposing a framework that jointly performs scene reconstruction and semantic completion incrementally in real-time, achieving accurate results.
Real-time scene reconstruction from depth data inevitably suffers from occlusion, thus leading to incomplete 3D models. Partial reconstructions, in turn, limit the performance of algorithms that leverage them for applications in the context of, e.g., augmented reality, robotic navigation, and 3D mapping. Most methods address this issue by predicting the missing geometry as an offline optimization, thus being incompatible with real-time applications. We propose a framework that ameliorates this issue by performing scene reconstruction and semantic scene completion jointly in an incremental and real-time manner, based on an input sequence of depth maps. Our framework relies on a novel neural architecture designed to process occupancy maps and leverages voxel states to accurately and efficiently fuse semantic completion with the 3D global model. We evaluate the proposed approach quantitatively and qualitatively, demonstrating that our method can obtain accurate 3D semantic scene completion in real-time.