HybridNet: Indoor segmentation with range and voxel fusion

In this paper, we propose a HybridNet that improves performance by fusing 2D and 3D features. A voxel-based method and a projection-based method were adopted to derive the results through one scan. Our approach consists of two parallel networks, extracts features along each dimension, and converges them in a Fusion Network. In the fusion network, the voxel blocks and 2D feature maps extracted from each structure are fused to the voxel grid and then trained through convolution. For effective training of 2D networks, we use data augmentation techniques using coordinate system rotation transformation. In addition, the performance was effectively improved by using a multi-loss with weights applied to each dimension, and better performance was achieved than the result using a single loss. Our proposed method can achieve better performance by changing the performance of the 2D network and 3D network, which can be generalized using other structures.