Feature Pyramid Network Based Efficient Normal Estimation and Filtering for Time-of-Flight Depth Cameras

Abstract

In this paper, an efficient normal estimation and filtering method for depth images acquired by Time-of-Flight (ToF) cameras is proposed. The method is based on a common feature pyramid networks (FPN) architecture. The normal estimation method is called ToFNest, and the filtering method ToFClean. Both of these low-level 3D point cloud processing methods start from the 2D depth images, projecting the measured data into the 3D space and computing a task-specific loss function. Despite the simplicity, the methods prove to be efficient in terms of robustness and runtime. In order to validate the methods, extensive evaluations on public and custom datasets were performed. Compared with the state-of-the-art methods, the ToFNest and ToFClean algorithms are faster by an order of magnitude without losing precision on public datasets.

Segmentation Methods Evaluation on Grapevine Leaf Diseases

“…3) Feature Pyramid Network: The Feature Pyramid Network FPN [15] architecture stands as a middle-ground between the lightness of MobileNetV3 and the accuracy of Mask R-CNN. We have seen the FPNs perform decently in surface normal estimation application [18], and canopy segmentation [19], since different support sizes are analogous on some levels to vine leaves. The base for implementing this method can be found at the link 4 .…”

MPI Planar Correction of Pulse Based ToF Cameras

“…The input size of this network can be arbitrary (depending on the internal parameters of the ToF camera) and the output results in a proportionally sized feature map with the input image using a fully convolutional approach. The approach is generic in the sense that for the convolutional architecture custom variants can be adopted [14] such as depth or IR images. The construction of these pyramids involves a bottomup and top-down path with lateral connections [12].…”