Voting for Voting in Online Point Cloud Object Detection
Open accessAbstract
This paper proposes an efficient and effective scheme to applying the sliding window approach popular in computer vision to 3D data.Specifically, the sparse nature of the problem is exploited via a voting scheme to enable a search through all putative object locations at any orientation.We prove that this voting scheme is mathematically equivalent to a convolution on a sparse feature grid and thus enables the processing, in full 3D, of any point cloud irrespective of the number of vantage points required to construct it.As such it is versatile enough to operate on data from popular 3D laser scanners such as a Velodyne as well as on 3D data obtained from increasingly popular push-broom configurations.Our approach is "embarrassingly parallelisable" and capable of processing a point cloud containing over 100K points at eight orientations in less than 0.5s.For the object classes car, pedestrian and bicyclist the resulting detector achieves best-in-class detection and timing performance relative to prior art on the KITTI dataset as well as compared to another existing 3D object detection approach.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The first convincing results for point cloud-only 3D bounding box estimation were produced by VoxelNet.
“The first convincing results for point cloud-only 3D bounding box estimation were produced by VoxelNet [167].”
From A Survey of Autonomous Driving: <i>Common Practices and Emerging Technologies</i> (2020) · cites this paper as [167]
The cited 2015 paper does not identify VoxelNet as producing the first convincing results. Its Related Works section says its own detector operates solely on 3D data, and its Evaluation section describes 3D bounding-box annotations and reports superior detection performance. Those passages run contrary to attributing the first such results to VoxelNet. (LLM, conf=0.86)
Checked by a reasoning model — may be wrong.
Experimental — accuracy not yet measuredVoxelNet's voxel feature encoder uses a fully connected neural network to convert the variable number of points in each occupied voxel to a feature vector of fixed size.
“Their voxel feature encoder (VFE) uses a fully connected neural network to convert the variable number of points in each occupied voxel to a feature vector of fixed size.”
From A Survey of Autonomous Driving: <i>Common Practices and Emerging Technologies</i> (2020) · cites this paper as [167]
The Overview says this paper maps points in each occupied cell to a fixed-dimensional feature vector. But its Feature Extraction section describes six computed features, not a fully connected neural network, and the copy does not describe VoxelNet’s voxel feature encoder. (LLM, conf=0.85)
Experimental — accuracy not yet measuredThe voxel grid encoded with feature vectors was then used as input to an RPN for multi-class object detection.
“The voxel grid encoded with feature vectors was then used as input to an aforementioned RPN for multi-class object detection.”
From A Survey of Autonomous Driving: <i>Common Practices and Emerging Technologies</i> (2020) · cites this paper as [167]
The Introduction describes converting the point cloud into a voxel feature grid, and Detection Performance reports results for three object classes. But Sections III–IV describe a sliding-window detector using a linear classifier, not a region proposal network (RPN). (LLM, conf=0.88)
Experimental — accuracy not yet measuredVoxelNet learned an encoding from raw point cloud data to voxel grid instead of hand-crafting input features computed during discretization.
“Instead of hand-crafting input features computed during the discretization process, VoxelNet learned an encoding from raw point cloud data to voxel grid.”
From A Survey of Autonomous Driving: <i>Common Practices and Emerging Technologies</i> (2020) · cites this paper as [167]
The Introduction and Section III describe this paper’s conversion of a point cloud into a grid of feature vectors. A search for VoxelNet and learned voxel encodings found no passage addressing whether VoxelNet learned its encoding instead of using hand-crafted input features. (LLM, conf=0.76)
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
Histograms of Oriented Gradients for Human Detection
Dalal, N., & Triggs, B. (2005) · 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05) · doi:10.1109/cvpr.2005.177 · Cited by 33,584 · see citing works
Vision meets robotics: The KITTI dataset
Geiger, A., Lenz, P., Stiller, C., & Urtasun, R. (2013) · The International Journal of Robotics Research · doi:10.1177/0278364913491297 · Cited by 9,913 · see citing works
Object Detection with Discriminatively Trained Part-Based Models
Felzenszwalb, P. F., Girshick, R. B., McAllester, D., & Ramanan, D. (2010) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2009.167 · Cited by 9,853 · see citing works
Flagged first, then by citations
Research works printed but not linked: 16
- J. Behley, V. Steinhage, and A.B. Cremers. Laser-based segment classification using a mixture of bag-of-words. In Intelligent Robots and Systems (IROS), 2013 IEEE/RSJ International Conference on, pages 4195-4200, Nov 2013.
- C. Dubout and F. Fleuret. Exact Acceleration of Linear Object Detectors. In Proceedings of the European Conference on Computer Vision (ECCV), pages 301-311, 2012.
- M. Everingham, L. Van Gool, C. K. I. Williams, J. Winn, and A. Zisserman. The Pascal Visual Object Classes (VOC) Challenge. International Journal of Computer Vision, 88(2):303-338, June 2010. · DOI · Paper page
- Sanja Fidler, Sven Dickinson, and Raquel Urtasun. 3D Object Detection and Viewpoint Estimation with a De- formable 3D Cuboid Model. In F. Pereira, C.J.C. Burges, L. Bottou, and K.Q. Weinberger, editors, Advances in Neural Information Processing Systems 25, pages 611- 619.
- Varsha Hedau, Derek Hoiem, and David Forsyth. Think- ing Inside the Box: Using Appearance Models and Context Based on Room Geometry. In Proceedings of the 11th European Conference on Computer Vision: Part VI, ECCV'10, pages 224-237, Berlin, Heidelberg, 2010. Springer-Verlag. ISBN 3-642-15566-9, 978-3-642- 15566-6.
- K. Lai, Liefeng Bo, Xiaofeng Ren, and D. Fox. Detection-based object labeling in 3D scenes. In Robotics and Automation (ICRA), 2012 IEEE International Con- ference on, pages 1330-1337, May 2012.
- Kevin Lai, Liefeng Bo, Xiaofeng Ren, and Dieter Fox. Sparse distance learning for object recognition com- bining RGB and depth information. In Robotics and Automation (ICRA), 2011 IEEE International Conference on, pages 4007 -4013, may 2011.
- Alain Lehmann, Bastian Leibe, and Luc Van Gool. Fast PRISM: Branch and Bound Hough Transform for Object Class Detection. International Journal of Computer Vision, 94(2):175-197, 2011. ISSN 0920-5691.
- A Neubeck and L. Van Gool. Efficient Non-Maximum Suppression. In Pattern Recognition, 2006. ICPR 2006. 18th International Conference on, volume 3, pages 850- 855, 2006.
- C. Premebida, J. Carreira, J. Batista, and U. Nunes. Pedestrian Detection Combining RGB and Dense LIDAR Data. In IROS, 2014.
- M. Quigley, Siddharth Batra, S. Gould, E. Klingbeil, Quoc Le, Ashley Wellman, and AY. Ng. High-accuracy 3D sensing for mobile manipulation: Improving object detection and door opening. In Robotics and Automation, 2009. ICRA '09. IEEE International Conference on, pages 2816-2822, May 2009.
- Shuran Song and Jianxiong Xiao. Sliding Shapes for 3D Object Detection in Depth Images. In David Fleet, Tomas Pajdla, Bernt Schiele, and Tinne Tuytelaars, editors, Computer Vision -ECCV 2014, volume 8694 of Lecture Notes in Computer Science, pages 634-651. Springer International Publishing, 2014. ISBN 978-3-319-10598- 7. · DOI
- Kah-Kay Sung and Tomaso Poggio. Example-Based Learning for View-Based Human Face Detection. IEEE Trans. Pattern Anal. Mach. Intell., 20(1):39-51, January 1998. ISSN 0162-8828.
- Alex Teichman and Sebastian Thrun. Tracking-Based Semi-Supervised Learning. In Proceedings of Robotics: Science and Systems, Los Angeles, CA, USA, June 2011.
- Alex Teichman, Jesse Levinson, and Sebastian Thrun. Towards 3D object recognition via classification of arbi- trary object tracks. In Robotics and Automation (ICRA), 2011 IEEE International Conference on, pages 4034 - 4041, may 2011.
- D.Z. Wang, I. Posner, and P. Newman. What Could Move? Finding Cars, Pedestrians and Bicyclists in 3D Laser Data. In Robotics and Automation (ICRA), 2012 IEEE International Conference on, may 2012.
Cited by
Sorted by citations
VoxelNet: End-to-End Learning for Point Cloud Based 3D Object Detection
Zhou, Y., & Tuzel, O. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00472 · Cited by 4,813 · see citing works
Multi-view 3D Object Detection Network for Autonomous Driving
Chen, X., Ma, H., Wan, J., Li, B., & Xia, T. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.691 · Cited by 3,450 · see citing works
Frustum PointNets for 3D Object Detection from RGB-D Data
Qi, C. R., Liu, W., Wu, C., Su, H., & Guibas, L. J. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00102 · Cited by 2,655 · see citing works
A Survey of Autonomous Driving: <i>Common Practices and Emerging Technologies</i>
Yurtsever, E., Lambert, J., Carballo, A., & Takeda, K. (2020) · IEEE Access · doi:10.1109/access.2020.2983149 · Cited by 1,889 · see citing works
“Shortly thereafter, a similar approach was used on point clouds created by lidars [163].”
Method3D Semantic Segmentation with Submanifold Sparse Convolutional Networks
Graham, B., Engelcke, M., & Maaten, L. V. D. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00961 · Cited by 1,788 · see citing works
PIXOR: Real-time 3D Object Detection from Point Clouds
Yang, B., Luo, W., & Urtasun, R. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00798 · Cited by 1,249 · see citing works
Vehicle Detection from 3D Lidar Using Fully Convolutional Network
Li, B., Zhang, T., & Xia, T. (2016) · Robotics: Science and Systems XII · doi:10.15607/rss.2016.xii.042 · Cited by 577 · see citing works
VoxelNeXt: Fully Sparse VoxelNet for 3D Object Detection and Tracking
Chen, Y., Liu, J., Zhang, X., Qi, X., & Jia, J. (2023) · arXiv · doi:10.48550/arxiv.2303.11301 · Cited by 486 · see citing works
HVNet: Hybrid Voxel Network for LiDAR Based 3D Object Detection
Ye, M., Xu, S., & Cao, T. (2020) · 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr42600.2020.00170 · Cited by 242 · see citing works
Multimodal vehicle detection: fusing 3D-LIDAR and color camera data
Asvadi, A., Garrote, L., Premebida, C., Peixoto, P., & J. Nunes, U. (2018) · Pattern Recognition Letters · doi:10.1016/j.patrec.2017.09.038 · Cited by 225 · see citing works
Deep semantic classification for 3D LiDAR data
Dewan, A., Oliveira, G. L., & Burgard, W. (2017) · 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros.2017.8206198 · Cited by 60 · see citing works
Multimodal Fusion Object Detection System for Autonomous Vehicles
Person, M., Jensen, M., Smith, A. O., & Gutierrez, H. (2019) · Journal of Dynamic Systems, Measurement, and Control · doi:10.1115/1.4043222 · Cited by 39 · see citing works
- Authors
- Zeng Wang, D., & Posner, I.
- Affiliations
- University of Oxford (GB)
- ORCID
- Posner, Ingmar
- Venue
- Robotics: Science and Systems XI
- Published
- 2015-07-13
- OpenAlex
- W2293349265
- Cited by
- 337
- References
- 3
- Citations by year
- 2026: 3 · 2025: 8 · 2024: 14 · 2023: 32 · 2022: 36 · 2021: 58 · 2020: 61 · 2019: 54 · 2018: 34 · 2017: 27 · 2016: 9 · 2015: 1
- FWCI
- 9.71
- Type
- conference paper
- Language
- en
- Open access
- bronze
- doi.org
- Locations
- Robotics: Science and Systems (publishedVersion)
- Funders
- Engineering and Physical Sciences Research Council (EP/J012017/1, EP/I005021/1)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Physical Sciences › Engineering › Computational Mechanics › 3D Shape Modeling and Analysis
- Physical Sciences › Engineering › Automotive Engineering › Autonomous Vehicle Technology and Safety
- Keywords
- 3D object detection, point clouds, voting scheme, sliding window, convolution, 3D laser scanner, parallel computing, KITTI dataset
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
