OverlapPredator

prs-eth/OverlapPredator is ranked 16,209 among these 8 peers; princeton-nlp/WebShop is ranked 16,208.

PREDATOR: Registration of 3D Point Clouds with Low Overlap (CVPR 2021, Oral) This repository represents the official implementation of the paper: PREDATOR: Registration of 3D Point Clouds with Low Overlap \*Shengyu Huang, \*Zan Gojcic, Mikhail Usvyatsov, Andreas Wieser, Konrad Schindler\ |ETH Zurich | \* Equal contribution For implementation using MinkowskiEngine backbone, please check this For more information, please see the project website Contact If you have any questions, please let us know: Shengyu Huang {shengyu. ch} News 2021-08-09: We've updated

arxiv version of our paper with improved performance! 2021-06-02: Fix feature gathering bug in k-nn graph, please see improved performance in this issue. Stay tunned for updates on other experiments! 2021-05-31: Check our video and poster on project page! 0, GeForce RTX 3090/GeForce GTX 1080Ti **Note**: We observe random data loader crashes due to memory issues, if you observe similar issues, please consider reducing the number of workers or increasing CPU RAM. We now

released a sparse convolution-based Predator, have a look here! Requirements To create a virtual environment and install the required dependences please run: in your working folder. Datasets and pretrained models For KITTI dataset, please follow the instruction on KITTI Odometry website to download the KITTI odometry training set. We provide preprocessed 3DMatch pairwise datasets (voxel-grid subsampled fragments together with their ground truth transformation matrices) raw dense 3DMatch datasets modelnet dataset pretrained models on 3DMatch, KITTI and Modelnet The

preprocessed data and models can be downloaded by running: To download raw dense 3DMatch data, please run: The folder is organised as follows: 3DMatch(Indoor) Train After creating the virtual environment and downloading the datasets, Predator can be trained using: Evaluate For 3DMatch, to reproduce Table 2 in our main paper, we first extract features and overlap/matachability scores by running: the features together with scores will be saved to . The estimation of the transformation parameters using RANSAC can then

be carried out using: dependent on used by RANSAC, this might take a few minutes. To evaluate PREDATOR on 3DLoMatch benchmark, please also change to in . Demo We prepared a small demo, which demonstrates the whole Predator pipeline using two random fragments from the 3DMatch dataset. To carry out the demo, please run: The demo script will visualize input point clouds, inferred overlap regions, and point cloud aligned with the estimated transformation parameters: ModelNet(Synthetic) Train To train PREDATOR on

ModelNet, please run: We provide a small script to evaluate Predator on ModelNet test set, please run: The rotation and translation errors could be better/worse than the reported ones due to randomness in RANSAC. KITTI(Outdoor) We provide a small script to evaluate Predator on KITTI test set, after configuring KITTI dataset, please run: the results will be saved to the log file. Custom dataset We have a few tips for train/test on custom dataset If it's similar

indoor scenes, please run first to check the generalisation ability before retraining Remember to voxel-downsample the data in your