Shape-Net: Room Layout Estimation from Panoramic Images Robust to Occlusion using Knowledge Distillation with 3D Shapes as Additional Inputs
This work addresses occlusion issues in room layout estimation for applications like virtual/augmented reality and furniture simulation, representing an incremental improvement over prior methods.
The paper tackles the problem of room layout estimation from panoramic images, which is hindered by occlusion, by proposing Shape-Net that uses knowledge distillation with 3D shapes as additional inputs to improve robustness. It achieves state-of-the-art performance on benchmark datasets and significantly improves accuracy on occluded images compared to existing models.
Estimating the layout of a room from a single-shot panoramic image is important in virtual/augmented reality and furniture layout simulation. This involves identifying three-dimensional (3D) geometry, such as the location of corners and boundaries, and performing 3D reconstruction. However, occlusion is a common issue that can negatively impact room layout estimation, and this has not been thoroughly studied to date. It is possible to obtain 3D shape information of rooms as drawings of buildings and coordinates of corners from image datasets, thus we propose providing both 2D panoramic and 3D information to a model to effectively deal with occlusion. However, simply feeding 3D information to a model is not sufficient to utilize the shape information for an occluded area. Therefore, we improve the model by introducing 3D Intersection over Union (IoU) loss to effectively use 3D information. In some cases, drawings are not available or the construction deviates from a drawing. Considering such practical cases, we propose a method for distilling knowledge from a model trained with both images and 3D information to a model that takes only images as input. The proposed model, which is called Shape-Net, achieves state-of-the-art (SOTA) performance on benchmark datasets. We also confirmed its effectiveness in dealing with occlusion through significantly improved accuracy on images with occlusion compared with existing models.