Enforcing geometric constraints of virtual normal for depth prediction

Abstract

Monocular depth prediction plays a crucial role in understanding 3D scene geometry. Although recent methods have achieved impressive progress in evaluation metrics such as the pixel-wise relative error, most methods neglect the geometric constraints in the 3D space. In this work, we show the importance of the high-order 3D geometric constraints for depth prediction. By designing a loss term that enforces one simple type of geometric constraints, namely, virtual normal directions determined by randomly sampled three points in the reconstructed 3D space, we can considerably improve the depth prediction accuracy. Significantly, the byproduct of this predicted depth being sufficiently accurate is that we are now able to recover good 3D structures of the scene such as the point cloud and surface normal directly from the depth, eliminating the necessity of training new sub-models as was previously done. Experiments on two benchmarks: NYU Depth-V2 and KITTI demonstrate the effectiveness of our method and state-of-the-art performance.

Fixed typos. Appearing in Proc. Int. Conf. Computer Vision 2019. Code is available at: https://tinyurl.com/virtualnormal

References in corpus (3)

• Depth Map Prediction from a Single Image using a Multi-Scale Deep Network
• DepthTransfer: Depth Extraction from Video Using Non-parametric Sampling
• Rethinking Monocular Depth Estimation with Adversarial Training

Cited by in corpus (5)

• Co-training for Deep Object Detection: Comparing Single-modal and Multi-modal Approaches
• Instance and Panoptic Segmentation Using Conditional Convolutions
• Normal Assisted Stereo Depth Estimation
• Balanced Depth Completion between Dense Depth Inference and Sparse Range Measurements via KISS-GP
• Polarimetric Monocular Dense Mapping Using Relative Deep Depth Prior