VoxNet: A 3D Convolutional Neural Network for real-time object recognition

By Daniel Maturana, Sebastian A. Scherer. 2015

3,723 citations recorded by the source corpus.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition

How researchers see this paper

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

In Aerospace Engineering

In Computer Vision and Pattern Recognition

Citations in the corpus

Learning Descriptor Networks for 3D Shape Synthesis and Analysis (2018, methodology): “From the perspective of statistical modeling, the existing 3D models can be grouped into two main categories: (1) 3D discriminators, such as Voxnet [16], which aim to learn a mapping from 3D voxel input to semantic labels for the purpose of 3D object clas-siﬁcation and recognition, and (2) 3D…”
PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation (2016, background): “Multiview CNNs: [24, 19] have tried to render 3D point cloud or shapes into 2D images and then apply 2D conv nets to classify them.”
3D Instance Segmentation of MVS Buildings (2021, methodology): “Volumetric-based methods [3], [4], and point cloud-based methods [5], [6] are limited by memory and computing power, thus are mainly used for the segmentation of relatively small indoor scenes [7], [8], [9], [10], [11], [12].”
VoxelNet: End-to-End Learning for Point Cloud Based 3D Object Detection (2017, methodology): “Some of the earlier methods for feature representation are [41, 8, 7, 19, 42, 35, 6, 27, 1, 36, 2, 25, 26].”
Occupancy Networks: Learning 3D Reconstruction in Function Space (2018, methodology): “Voxel Representations: Due to their simplicity, voxels are the most commonly used representation for discriminative [45,55,63] and generative [9,23,58,64,75,77] 3D tasks.”
A Point Set Generation Network for 3D Object Reconstruction from a Single Image (2016, methodology): “t image Xwill be augmented by a random variable that is conditioned on Y, which takes the volumetric representation of the groundtruth shape S. A 3D convolutional network is used as the encoder Q(see [16] for a good reference of 3D conv networks). Therefore, a local proximity in the embedding space ”
RealPoint3D: Point Cloud Generation from a Single Image with Complex Background (2018, methodology): “(Maturana and Scherer 2015) and (Qi et al. 2016) are two pioneers of using voxel-based methods for object detection and classiﬁcation.”
A Survey of Deep Network Solutions for Learning Control in Robotics: From Reinforcement to Imitation (2016, neutral): “D DNN Benchmark X Object Detection Grasping FF F Redmon and Angelova (2015) RGBD CNN Benchmark X Object Detection Grasping F FF Li et al. (2016) 3D Lidar CNN Benchmark Object Detection Auto Drive F F Maturana and Scherer (2015a) Lidar 3D Conv Self-Labeled X Classiﬁcation Aerial FF FF Engelcke et al.”
Revisiting Point Cloud Classification: A New Benchmark Dataset and Classification Model on Real-World Data (2019, methodology): “, using multiple view images [39, 48, 46, 22], or applying convolutions on 3D voxel grids [27, 43].”
SyncSpecCNN: Synchronized Spectral CNN for 3D Shape Segmentation (2016, methodology): “In addition we design an additional baseline using a 3D volumetric CNN architecture, denoted as Voxel CNN, which generalizes VoxNet [19] for segmentation tasks.”
OctNet: Learning Deep 3D Representations at High Resolutions (2016, methodology): “[29] which operates on a fixed 32(3) voxel grid, the proposed OctNet and a dense version of it which we denote “DenseNet” in the following.”
Volumetric and Multi-view CNNs for Object Classification on 3D Data (2016, methodology): “A similar approach is VoxNet [21], which also uses binary voxel grids and a corresponding 3D CNN architecture.”
Dynamic Graph CNN for Learning on Point Clouds (2018, methodology): “++. An advanced version including a local-aware network and dynamical graph recomputation achieves best results on this dataset. MEAN OVERALL CLASS ACCURACY ACCURACY 3DSHAPENETS [54] 77.3 84.7 VOXNET [30] 83.0 85.9 SUBVOLUME [35] 86.0 89.2 ECC [45] 83.2 87.4 POINTNET [34] 86.0 89.2 POINTNET++ [36] -”
Dynamic Edge-Conditioned Filters in Convolutional Neural Networks on Graphs (2017, neutral): “4) against two methods based on volumetric CNNs evaluated on voxelized occupancy grids of size 32x32x32 (VoxNet [25] 73.”
3D fully convolutional network for vehicle detection in point cloud (2016, methodology): “hannel of value f0;1gis used to present whether there is any points observed at the corresponding grid elements. Some more sophisticated features have also been introduced in the previous works, e.g. [20]. The mechanism of 2D CNN naturally extends to 3D on the square grids. Figure 2 shows an example”
Vote3Deep: Fast object detection in 3D point clouds using efficient convolutional neural networks (2016, background): “Dense 3D occupancy grids obtained from point clouds are processed with CNNs in [8] and [9].”
Vehicle Detection from 3D Lidar Using Fully Convolutional Network (2016, methodology): “For 3D range scan some works discretize point cloud along 3D grids and train 3D CNN structure for classiﬁcation [33, 19].”
Deep Sliding Shapes for Amodal 3D Object Detection in RGB-D Images (2015, background): “Several recent works [17, 5, 31, 13] also extract deep learning features for retrieval and classification of CAD models.”
Multi-View Silhouette and Depth Decomposition for High Resolution 3D Object Representation (2018, background): “Deep Learning with 3D Data Recent advances with 3D data have leveraged deep learning, beginning with architectures such as 3D convolutions [25, 19] for object classification.”