Performance of global descriptors for velodyne-based urban object recognition

By Tongtong Chen, Bin Dai, Daxue Liu, Jinze Song. 2014

31 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Dynamic Edge-Conditioned Filters in Convolutional Neural Networks on Graphs (2017, neutral): “…) denotes ECC with c output channels followed by afﬁne batch normalization and ReLU activation, MP( r , ρ ) stands for max-pooling down to grid resolution of r meters and neighborhood radius of Model Mean F1 Triangle+SVM [9] 67.1 GFH+SVM [7] 71.0 VoxNet [26] 73.0 ORION [1] 77.8 ECC 2 ρ 74.4 ECC 1 .”
Orientation-boosted Voxel Nets for 3D Object Recognition (2016, neutral): “Triangle+SVM [17] 24 59 37 81 96 57 64 57 62 68 31 40 30 50 GFH+SVM [24] 0 81 55 82 94 68 73 63 74 73 25 52 19 59 Our Baseline 48 63 59 85 97 68 57 75 66 82 32 52 48 59 ORIO Basic 42 79 63 85 98 73 68 69 66 84 40 61 51 59 ORION Fusion 46 76 64 86 98 75 72 73 68 81 46 62 50 66 ORION F2C 43 78 64 85 9”
3DmFV: Three-Dimensional Point Cloud Classification in Real-Time Using Convolutional Neural Networks (2018, methodology): “Some descriptors were specifically designed for the task of object classification in outdoor LiDAR data [15], [16].”
VoxNet: A 3D Convolutional Neural Network for real-time object recognition (2015, methodology): “Table III shows our best VoxNet against the best approach from [21], which combines an unsupervised form of Deep Learning with SVM classifiers, and [37], which designs a rotationally invariant descriptor and classifies it with a nonlinear SVM.”
General-Purpose Deep Point Cloud Feature Extractor (2018, unknown): “1 GFH+SVM[12] 71.”
Lidar for Autonomous Driving: The Principles, Challenges, and Trends for Automotive Lidar and Perception Systems (2020, methodology): “From the arsenal of ML, plenty of ML tools such as naive Bayes [25], support vector machines (SVMs) [21], [24], [28], k-nearest neighbor (NN) [26], random forest (RF), and gradient boosting trees [29] can be applied.”