The task of predicting smooth and edge-consistent depth maps is notoriously difficult for single image depth estimation. This paper proposes a novel Bilateral Grid based 3D convolutional neural network, dubbed as 3DBG-UNet, that parameterizes high dimensional feature space by encoding compact 3D bilateral grids with UNets and infers sharp geometric layout of the scene. Further, another novel 3DBGES-UNet model is introduced that integrate 3DBG-UNet for inferring an accurate depth map given a single color view. The 3DBGES-UNet concatenates 3DBG-UNet geometry map with the inception network edge accentuation map and a spatial object's boundary map obtained by leveraging semantic segmentation and train the UNet model with ResNet backbone. Both models are designed with a particular attention to explicitly account for edges or minute details. Preserving sharp discontinuities at depth edges is critical for many applications such as realistic integration of virtual objects in AR video or occlusion-aware view synthesis for 3D display applications.The proposed depth prediction network achieves state-of-the-art performance in both qualitative and quantitative evaluations on the challenging NYUv2-Depth data. The code and corresponding pre-trained weights will be made publicly available.
翻译:在单一图像深度估计中,预测光滑和地缘一致的深度地图的任务非常困难。 本文提议了一个新的基于3DBG- UNet的双边3D进化神经网络, 称为3DBG- UNet, 该网络通过与UNets编码紧凑的3D双边网格, 并推断出场景的精确几何布局, 将高维地貌空间参数化为参数。 此外, 还引入了另一个新的3DBG-UNet模型, 将3DBG- UNet 的精确深度地图结合成一个单一颜色视图, 以推断出准确的深度地图。 3DBG- UNet 3DG- Uet 进化神经网络, 以3DBG- UNet 星座图和空间天体的边界地图, 以 3DBG- 3DG- Unet 星座图为首, 3DBG- 3DG-UNet 地理测量图, 3DG- 3DG-UNet 的地图, 3DG- 和 UNet Get 地理测量图, 地理图由利用 3DG- 3DG- 3DG- UDG- Unew concalbet 的边界图的边界图的边界图, 和 地图图, 和空间天体图图图图, 和图图地图, 3DG- 3DG- 和 3DG- grouttrouttroute- get 3DG- get 地理图, 3DG- get 和 3DG- get 地图, 3DG- 3DG- get 3DG- get 地理图, 地理图, 地理图地图, 地理图地图, 3DG- 地理图图图图图图图图地图, 3DB 3G- get 3DG- get 3DG- get 地图, 地理图, 3DG- get 3DG- get 3DG- get 3DG- get 3DG- get