RESUMO
In this paper, we propose a method to refine the depth maps obtained by Multi-View Stereo (MVS) through iterative optimization of the Neural Radiance Field (NeRF). MVS accurately estimates the depths on object surfaces, and NeRF accurately estimates the depths at object boundaries. The key ideas of the proposed method are to combine MVS and NeRF to utilize the advantages of both in depth map estimation and to use NeRF for depth map refinement. We also introduce a Huber loss into the NeRF optimization to improve the accuracy of the depth map refinement, where the Huber loss reduces the estimation error in the radiance fields by placing constraints on errors larger than a threshold. Through a set of experiments using the Redwood-3dscan dataset and the DTU dataset, which are public datasets consisting of multi-view images, we demonstrate the effectiveness of the proposed method compared to conventional methods: COLMAP, NeRF, and DS-NeRF.
RESUMO
It is necessary to estimate the pose of the probe with high accuracy to reconstruct 3D ultrasound (US) images only from US image sequences scanned by a 1D-array probe. We propose the probe pose estimation method using Convolutional Neural Network (CNN) with training by image reconstruction loss. To calculate the image reconstruction loss, we use the image reconstruction network which consists of an encoder that extracts features from the two US images and a decoder that reconstructs the intermediate US image between the two images. CNN is trained to minimize the image reconstruction loss between the ground-truth image and the reconstructed image. Through experiments, we demonstrate that the proposed method exhibits efficient performance compared with the conventional methods.