Kangcheng Liu Jimmy's starred repositories
CenterTrack
Simultaneous object detection and tracking using center points.
libpointmatcher
An Iterative Closest Point (ICP) library for 2D and 3D mapping in Robotics
kitti_object_vis
KITTI Object Visualization (Birdview, Volumetric LiDar point cloud )
awesome-local-global-descriptor
My personal note about local and global descriptor
FewShot_GAN-Unet3D
Tensorflow implementation of our paper: Few-shot 3D Multi-modal Medical Image Segmentation using Generative Adversarial Learning
FewShotDetection
(ECCV 2020) PyTorch implementation of paper "Few-Shot Object Detection and Viewpoint Estimation for Objects in the Wild"
RePRI-for-Few-Shot-Segmentation
(CVPR 2021) Code for our method RePRI for Few-Shot Segmentation. Paper at http://arxiv.org/abs/2012.06166
HPLFlowNet
Code for our CVPR 2019 paper, HPLFlowNet: Hierarchical Permutohedral Lattice FlowNet for Scene Flow Estimation on Large-scale Point Clouds.
Point-Spatio-Temporal-Convolution
Implementation of the "PSTNet: Point Spatio-Temporal Convolution on Point Cloud Sequences" paper.
kitti-velodyne-viewer
view kitti lidar point cloud with bounding box label
DetectionTutorialSpaceNet
This is a tutorial on training a network to detect buildings with the SpaceNet data.
contactdb_utils
Python and ROS (C++) utilities for the ContactDB dataset
CS543_project_Image-based-Localization-of-Bridge-Defects-with-AR-Visualization
Visual inspection of bridges is customarily used to identify and evaluate faults. However, current procedures followed by human inspectors demand long inspection times to examine large and difficult to access bridges. To address these limitations, we investigate a computer vision‐based approach that employs SIFT keypoint matching on collected images of defects against a pre-existing reconstructed 3D point cloud of the bridge. We also investigate methods of reducing computation time with ML-based and conventional CV methods of segmentation to eliminate redundant keypoints. Our project successfully localizes the defect images and achieves a savings in runtime from filtering keypoints.
visualisation-perception-3D
Several tools for visualizing 3D perception result
BDCI2018-pointnet-
This is the source code for BDCI-2018 阿里巴巴自动驾驶三维点云语义分割 of our team (试一下pointnet)
GEOSLAM
The seismo-lineament analysis method is a tool to spatially correlate a shallow-focus earthquake to the surface trace of the fault that generated it. SLAM is the intellectual property and work product of Vince Cronin. The GEOSLAM Python code was a translation from Vince Cronin's Mathematica files, where this translation was performed by Luke Pajer.
Visualization_KITTI_Waymo
Displays the data for successive frames of Kitti and Waymo;Waymo data parsing is no longer declared here