Towards Balanced RGB-TSDF Fusion for Consistent Semantic Scene Completion by 3D RGB Feature Completion and a Classwise Entropy Loss Function
This work improves 3D scene understanding for applications like robotics and autonomous driving by enhancing fusion of color and geometry data, though it is incremental as it builds on existing RGB-TSDF fusion methods.
The paper tackles the problem of inconsistent results in RGB-TSDF fusion for Semantic Scene Completion by addressing the sparsity of RGB features in 3D space, achieving state-of-the-art performance on public datasets without extra data.
Semantic Scene Completion (SSC) aims to jointly infer semantics and occupancies of 3D scenes. Truncated Signed Distance Function (TSDF), a 3D encoding of depth, has been a common input for SSC. Furthermore, RGB-TSDF fusion, seems promising since these two modalities provide color and geometry information, respectively. Nevertheless, RGB-TSDF fusion has been considered nontrivial and commonly-used naive addition will result in inconsistent results. We argue that the inconsistency comes from the sparsity of RGB features upon projecting into 3D space, while TSDF features are dense, leading to imbalanced feature maps when summed up. To address this RGB-TSDF distribution difference, we propose a two-stage network with a 3D RGB feature completion module that completes RGB features with meaningful values for occluded areas. Moreover, we propose an effective classwise entropy loss function to punish inconsistency. Extensive experiments on public datasets verify that our method achieves state-of-the-art performance among methods that do not adopt extra data.