2019 · 151 citations · 30 references
ComputingComputer ScienceRobotics EngineeringComputer VisionMachine LearningRobot Vision3D ReconstructionMulti-view GeometryPoint Cloud ProcessingPose EstimationRobotic VisionScene Understanding2D Pose Estimation3D Image Analysis3D Pose Estimation3D Scene UnderstandingPoint Cloud GenerationPoint Cloud Matching
Large-scale point cloud generated from 3D sensors is more accurate than its image-based counterpart. However, it is seldom used in visual pose estimation due to the difficulty in obtaining 2D-3D image to point cloud correspondences. In this paper, we propose the 2D3D-MatchNet - an end-to-end deep network architecture to jointly learn the descriptors for 2D and 3D keypoint from image and point cloud, respectively. As a result, we are able to directly match and establish 2D-3D correspondences from the query image and 3D point cloud reference map for visual pose estimation. We create our Oxford 2D-3D Patches dataset from the Oxford Robotcar dataset with the ground truth camera poses and 2D-3D image to point cloud correspondences for training and testing the deep network. Experimental results verify the feasibility of our approach.
30
ImageNet: A large-scale hierarchical image database
Jia Deng, Wei Dong, Richard Socher et al. · 2009 IEEE Conference on Computer Vision and Pattern Recognition · 2009 · 61.8K citations
Martin A. Fischler, Robert C. Bolles · Communications of the ACM · 1981 · 25.5K citations · Full text
FaceNet: A unified embedding for face recognition and clustering
2015 · 11.2K citations · Full text