Pith. sign in

REVIEW 3 major objections 6 minor 43 references

Fast Point R-CNN

T0 review · 3 major / 6 minor · reviewed 2026-08-14 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Fast Point R-CNN fuses voxel grids with raw point coordinates in a two-stage detector, reporting state-of-the-art KITTI 3D and BEV accuracy while running at 15 FPS.

desk verdict Solid two-stage LiDAR detector with careful ablations, but the 'state-of-the-art' claim is contradicted by its own Table 1. read the letter →

arxiv 1908.02990 v2 pith:UR7MFELN submitted 2019-08-08 cs.CV

Yilun Chen , Shu Liu , Xiaoyong Shen , Jiaya Jia This is my paper · ORCID
classification cs.CV
keywords 3DobjectdetectionpointcloudLiDARvoxelrepresentationtwo-stagedetectorattentionfeaturefusionKITTIbenchmarkreal-time
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper tries to show that a LiDAR-only detector can be both fast and accurate by exploiting two complementary point-cloud representations: voxel grids, which are cheap for convolutional networks to process, and raw point coordinates, which preserve precise localization. Its two-stage design first uses a lightweight convolutional network on voxels to emit a small number of high-quality proposals, then a refinement network inspects the raw points inside each proposal and fuses their exact coordinates with the voxel network's convolutional features through an attention mechanism. The authors claim this achieves state-of-the-art performance on the KITTI benchmark for both 3D and bird's-eye-view car detection, at 15 frames per second on a Tesla P40 GPU, using only LiDAR point clouds and no RGB images.

What carries the argument

The load-bearing mechanism is the attention-based fusion module inside RefinerNet. For each point inside a proposal (plus a 0.3 m margin), the convolutional feature map from VoxelRPN is indexed at the point's projected BEV location and concatenated with a 128-dimensional MLP embedding of the point's canonized coordinates; a sigmoid attention computed from the convolutional features gates this combined representation before two pointwise MLP layers and a max-pool aggregate the box into a fixed-size feature. This lets each raw point retain exact position while still receiving context and receptive-field information from the voxel convolutional stream. A second supporting mechanism is the canonized-corner regression target: points and ground-truth corners are rotated and translated into the proposal's coordinate frame, ordering the eight corners deterministically, and the network regresses offsets to those ordered corners rather than a direct box parameterization.

What would settle it

Run the full pipeline on the KITTI validation split with RefinerNet fed (a) VoxelRPN proposals and (b) a high-recall alternative (e.g., ground-truth proposals or a proposal set with deliberately lowered recall via weakened augmentation), and compare final AP0.7 and AP0.8 for both near and distant objects. If feeding higher-recall proposals does not substantially improve final AP, refinement is doing little beyond rescoring; if lowering VoxelRPN recall by a few points collapses final AP, the claimed two-stage benefit is contingent on first-stage recall being near-perfect.

Watch

Extended reading notes

Core claim

The central claim is that the speed-accuracy trade-off in 3D detection can be shifted by splitting the problem along representational lines rather than architectural lines. A voxel-based region proposal network (VoxelRPN), built from a small number of 3D convolutions followed by 2D convolutions, produces a compact set of initial boxes quickly and with high recall. The second stage (RefinerNet) then re-examines the raw points inside each proposed box, canonizes them by rotating and translating into the proposal frame, and fuses an MLP embedding of each point's coordinates with the convolutional feature from VoxelRPN at the point's projected location. A sigmoid attention gate applied to the convolutional feature emphasizes informative context before two pointwise MLP layers and a max-pool aggregate the box's points. This fusion restores the precise localization lost during voxelization and strided convolutions; the paper reports that the refinement raises moderate 3D AP0.7 from 76.64 to 79.00 on the KITTI validation split, with large gains at higher IoU thresholds (AP0.8 rising from 42.6 to 52.95) and for distant objects in the 30-50 m range.

Load-bearing premise

The central assumption is that the first-stage VoxelRPN already proposes essentially all true objects (over 95% recall at 0.5 BEV IoU within the top 30 boxes for cars), because RefinerNet can only re-score and re-shape boxes that the first stage finds; any object missed by VoxelRPN is missed by the whole system.

Editorial extensions

If this is right

  • If the central claim holds, high-quality 3D detection no longer requires RGB image input or multi-sensor fusion; LiDAR alone can match or exceed the accuracy of sensor-fusion methods while running faster.
  • Refining only a small set of proposals keeps the second stage cheap: the full pipeline runs at 0.065 s per frame on a Tesla P40, opening the door to real-time use in autonomous driving.
  • The refinement stage provides its largest gains precisely where it matters most: high-IoU metrics (AP0.8) and distant objects (30-50 m), which are the failure modes most relevant to safety.
  • The two-stage decoupling means future improvements to first-stage recall (more anchors, higher resolution, better augmentation) translate directly into final-AP gains, since the refinement stage only re-shapes boxes the first stage already finds.
  • The method's speed advantage is complementary to sparse-convolution and TensorRT-style engineering, suggesting further speedups are possible without sacrificing accuracy.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The coordinate-plus-context attention fusion could generalize beyond detection to other point-cloud tasks like semantic segmentation or tracking, where the same tension between precise geometry and contextual understanding exists.
  • Because RefinerNet only sees points inside the proposed box plus a small margin, the system's recall ceiling is set entirely by VoxelRPN; for pedestrian and cyclist classes, where the paper reports lower first-stage recall, a denser or scale-aware first stage would be the natural next step.
  • The reported 15 FPS figure is on a Tesla P40 without sparse convolution or TensorRT; combining these complementary acceleration techniques could push the method toward the speed of the fastest single-stage detectors while keeping the refinement gain.
  • The canonized-corner loss may be a reusable recipe for any 3D box regressor that must handle orientation ambiguity, since it removes the corner-ordering uncertainty that plagues direct corner regression.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 6 minor

Summary. The paper proposes Fast Point R-CNN, a two-stage LiDAR-based 3D object detector. The first stage, VoxelRPN, voxelizes the point cloud and uses a lightweight combination of 3D and 2D convolutions to produce a small set of high-recall proposals. The second stage, RefinerNet, crops raw points around each proposal, fuses point coordinates with convolutional features via an attention mechanism, and applies a light PointNet to refine the box corners using a canonized corner regression target. The method is evaluated on the KITTI benchmark for car, pedestrian, and cyclist detection, with ablations on the validation split. On the KITTI test set it reports 75.73 moderate 3D AP for cars at 15 FPS on a Tesla P40, and the authors claim state-of-the-art performance in both 3D and BEV detection.

Significance. If the technical claims hold, the architecture is a useful engineering contribution: it demonstrates that a voxel-based proposal stage plus a raw-point refinement stage can be competitive with much heavier point-wise detectors, and the ablation study gives clear evidence that each component—3D convolutions, higher resolution input, MIXUP augmentation, multi-angle anchors, coordinate features, and the attention fusion module—contributes to the final accuracy. The canonized corner loss and the analysis of distant-object improvement (Table 5) are informative. However, the headline claim of state-of-the-art accuracy is not supported by the paper's own Table 1, and the efficiency claim is presented without the hardware/software context needed for comparison. The absence of code or multiple-run statistics further limits verification. The work is nevertheless a solid, reproducible-in-principle two-stage detector that should be accepted if the claims are calibrated and the stated missing analyses are provided.

major comments (3)
  1. [Abstract and Section 4.2, Table 1] The assertion that Fast Point R-CNN "achieves state-of-the-arts" is not supported by the numbers in Table 1. On the KITTI primary metric (3D AP@0.7 moderate), the method scores 75.73, which is lower than PointRCNN's 75.76. In BEV moderate it ties PointPillars at 86.10, and on BEV hard it is lower (78.17 vs 79.83). There is no row in Table 1 in which the method is strictly ahead of all listed methods. Please revise the abstract and Section 4.2 to describe the results as competitive with state-of-the-art, or state precisely the metric and difficulty level for which a state-of-the-art claim is made.
  2. [Section 4.2 and Table 1] The speed claim needs to be framed more carefully. The paper says Fast Point R-CNN runs at 15 FPS, but Table 1 lists PointPillars at 0.016 s (about 62 FPS) and SECOND at 0.05 s (20 FPS). Direct comparison is complicated by different GPUs (Tesla P40 vs GTX 1080Ti) and by PointPillars' use of TensorRT, which the paper acknowledges. As written, however, the abstract's "state-of-the-arts with a 15FPS detection rate" implies a Pareto advantage in both accuracy and speed that Table 1 does not show. Please report the exact hardware and inference optimizations used for each method and avoid implying that the method is both the fastest and the most accurate in the comparison.
  3. [Section 3.4 and Section 5.3, Table 5] The two-stage design can only refine proposals produced by VoxelRPN, and the paper justifies this by stating that VoxelRPN recall exceeds 95% at 0.5 BEV IoU in the top 30 boxes for the car class. No recall numbers are given for pedestrians, cyclists, or distant objects. Table 5 shows that VoxelRPN's 3D AP0.7 at 30-50 m is only 51.99, and the fact that RefinerNet improves it to 58.41 does not tell us what fraction of distant objects were missed entirely by the first stage and therefore unrecoverable. Please report VoxelRPN recall by class and by distance range, and analyze how many missed proposals are recoverable by RefinerNet; this is necessary to support the general claim that the framework is effective beyond the car class in near-range settings.
minor comments (6)
  1. [Section 4.1, Implementation Details] The car anchor size is reported as ha=1.73, wa=0.6, la=0.8 meters, which is identical to the pedestrian anchor size and physically implausible for KITTI cars; please confirm the correct anchor dimensions and correct the text if this is a typo.
  2. [Section 4.1, Implementation Details] For pedestrian and cyclist, the text lists anchor sizes ha=1.73, wa=0.6, la=0.8 and ha=1.73, wa=0.6, la=1.76; please clarify whether the car anchor is actually the same as the pedestrian anchor and explain how these anchor sizes were chosen.
  3. [Figure 2] The notation "3×" and "5×" near the 2D convolution blocks is not explained in the caption or the text; please specify the number of repeated layers and the meaning of the stride tuples such as /(2,2,2).
  4. [Section 5.2, Table 4] The fusion baseline labeled "Concatenation" is not fully specified: it is unclear whether the coordinate feature and convolutional feature are concatenated before or after the PointNet MLPs and where the attention mask is applied in the proposed module. Please provide a precise description so the ablation is reproducible.
  5. [Section 5.2, RoI Align comparison] The rotated RoI align baseline is described as using 4096D MLP layers after pooling with size 8×4; please state whether the same training schedule, positive-proposal threshold, and data augmentation were used as in RefinerNet, so the comparison is apples-to-apples.
  6. [General] No code or pretrained models are provided, and the performance difference from PointRCNN on the primary metric is 0.03 AP; releasing code or at least detailed inference scripts would materially improve reproducibility and help readers verify the reported results.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the paper reports benchmark results on KITTI and does not rename fitted parameters as predictions.

full rationale

Fast Point R-CNN's central claims are accuracy and speed measurements obtained by training on KITTI and evaluating on the KITTI test server. The loss functions (Eqs. 1-6) define training objectives; they do not by construction determine the reported AP values, which are external benchmark measurements. Ablation studies (Tables 3-6) compare independently trained model variants, and no parameter fitted to a subset of the evaluation data is then presented as a prediction of closely related quantities. The only notable flaw is that the abstract's 'state-of-the-arts' claim is not supported by the paper's own Table 1 (PointRCNN achieves a higher 3D moderate AP), but that is a correctness or framing issue, not a circularity. Self-citations such as IPOD are not load-bearing for the method's derivation, and no uniqueness theorem or hidden ansatz is imported from prior work. Therefore the score is 0.

Assumptions & free parameters 6 free parameters · 3 assumptions · 0 invented entities

The central claim rests on hand-chosen hyperparameters (anchor sizes, voxel size, loss weights) and on the reliability of the KITTI benchmark and the stability of CNN training. No new physical entities are introduced.

free parameters (6)
  • Car anchor size (h, w, l) = 1.73, 0.6, 0.8 meters
    Chosen to match average car dimensions in KITTI; used in VoxelRPN anchor matching (Section 4.1).
  • Voxel size = 0.1 x 0.1 x 0.2 m
    Resolution of the voxel grid; hand-chosen tradeoff between speed and accuracy.
  • Points per voxel = 6
    The paper states a point-wise 8-channel MLP with 6 points per voxel is adequate; this is a hand-tuned capacity choice.
  • Class imbalance weight gamma = 10
    Used in the classification loss to weight negative samples; set empirically.
  • Smooth L1 sigma = 3
    Scale parameter in the regression loss.
  • MIXUP objects per frame = 20
    Number of ground-truth objects pasted into each training frame for augmentation.
assumptions (3)
  • domain assumption The KITTI dataset annotations and evaluation server are reliable indicators of real-world detection quality.
    The paper's claims rest entirely on KITTI AP numbers, which are computed by an external server.
  • domain assumption Random rotations and translations of ground-truth boxes during augmentation preserve the physical plausibility of the scene.
    The MIXUP and box-jitter augmentations assume pasted objects with small context regions are valid training samples; this is not tested for physical realism.
  • domain assumption The deep networks used converge to a good local optimum with the given hyperparameters.
    Stochastic optimization results are empirical; the paper does not prove convergence.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Fast Point R-CNN." pith.science (2026). https://pith.science/paper/UR7MFELN

@misc{pith2026190802990,
  author       = {Pith},
  title        = {Pith review of: Fast Point R-CNN},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/UR7MFELN}},
  note         = {Machine review of arXiv:1908.02990}
}
read the original abstract

We present a unified, efficient and effective framework for point-cloud based 3D object detection. Our two-stage approach utilizes both voxel representation and raw point cloud data to exploit respective advantages. The first stage network, with voxel representation as input, only consists of light convolutional operations, producing a small number of high-quality initial predictions. Coordinate and indexed convolutional feature of each point in initial prediction are effectively fused with the attention mechanism, preserving both accurate localization and context information. The second stage works on interior points with their fused feature for further refining the prediction. Our method is evaluated on KITTI dataset, in terms of both 3D and Bird's Eye View (BEV) detection, and achieves state-of-the-arts with a 15FPS detection rate.

Figures

Figures reproduced from arXiv: 1908.02990 by the authors.

Figure 1
Figure 1. Overview of our two-stage framework. In the first stage, we voxelize point cloud and feed them to VoxelRPN to produce a [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 3
Figure 3. Network Structure of RefinerNet. C M C C M (a) (c) (d) (b) Concatenation Element-wise Multiplication sigmoid Proposal GT 1 2 3 4 [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figure 4
Figure 4. Canonization of a box. The number denotes the order of [PITH_FULL_IMAGE:figures/full_fig_p004_4.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: ∆2xi = xi,g − xp, ∆2yi = yi,g − yp, ∆2zi = zi,g − zp (6) This parameterization is a general and natural design for RefinerNet that processes directly on coordinates of points. 4. Experiments We conduct experiments on the challenging KITTI [8] dataset in terms of 3D det…
Figure 5
Figure 5. Figure 5: Visualization of our results. Method Range (meters) 3D (Moderate) BEV (Moderate) AP0.7 AP0.8 AP0.7 AP0.8 VoxelRPN 0-30 88.39 58.81 90.22 83.32 Fast Point R-CNN 0-30 89.26 62.73 90.25 85.61 VoxelRPN 30-50 51.99 13.31 73.51 49.63 Fast Point R-CNN 30-50 58.41 15.39 73.9 5…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

43 extracted references · 27 canonical work pages

  1. [1]

    YOLO3D: End-to-end real-time 3D Oriented Object Bounding Box Detection from LiDAR Point Cloud

    Waleed Ali, Sherif Abdelkarim, Mohamed Zahran, Mah- moud Zidan, and Ahmad El Sallab. Yolo3d: End-to-end real-time 3d oriented object bounding box detection from li- dar point cloud. arXiv:1808.02350, 2018

    work page Pith review arXiv 2018

  2. [2]

    BirdNet: a 3D Object Detection Framework from LiDAR information

    Jorge Beltran, Carlos Guindel, Francisco Miguel Moreno, Daniel Cruzado, Fernando Garcia, and Arturo de la Escalera. Birdnet: a 3d object detection framework from lidar informa- tion. arXiv:1805.01195, 2018

    work page Pith review arXiv 2018

  3. [3]

    Monocular 3d object de- tection for autonomous driving

    Xiaozhi Chen, Kaustav Kundu, Ziyu Zhang, Huimin Ma, Sanja Fidler, and Raquel Urtasun. Monocular 3d object de- tection for autonomous driving. In CVPR, 2016

    work page 2016

  4. [4]

    3d object proposals for accurate object class detection

    Xiaozhi Chen, Kaustav Kundu, Yukun Zhu, Andrew G Berneshawi, Huimin Ma, Sanja Fidler, and Raquel Urtasun. 3d object proposals for accurate object class detection. In NIPS, 2015

    work page 2015

  5. [5]

    Multi-view 3d object detection network for autonomous driving

    Xiaozhi Chen, Huimin Ma, Ji Wan, Bo Li, and Tian Xia. Multi-view 3d object detection network for autonomous driving. In CVPR, 2017

    2017

  6. [6]

    Fully-convolutional point networks for large-scale point clouds

    Rethage Dario, Wald Johanna, Sturm Jrgen, Navab Nassir, and Tombari Federico. Fully-convolutional point networks for large-scale point clouds. In ECCV, 2018

    work page 2018

  7. [7]

    Cut, paste and learn: Surprisingly easy synthesis for instance de- tection

    Debidatta Dwibedi, Ishan Misra, and Martial Hebert. Cut, paste and learn: Surprisingly easy synthesis for instance de- tection. In ICCV, 2017

    2017

  8. [8]

    Are we ready for autonomous driving? the kitti vision benchmark suite

    Andreas Geiger, Philip Lenz, and Raquel Urtasun. Are we ready for autonomous driving? the kitti vision benchmark suite. In CVPR, 2012

    2012

Show all 43 references
  1. [9]

    Fast r-cnn

    Ross Girshick. Fast r-cnn. In ICCV, 2015

    2015

  2. [10]

    3d semantic segmentation with submanifold sparse convolutional networks

    Benjamin Graham, Martin Engelcke, and Laurens van der Maaten. 3d semantic segmentation with submanifold sparse convolutional networks. CVPR, 2018

    2018

  3. [11]

    Mask r-cnn

    Kaiming He, Georgia Gkioxari, Piotr Doll ´ar, and Ross Gir- shick. Mask r-cnn. In ICCV, 2017

    2017

  4. [12]

    Adam: A method for stochastic optimization

    Diederik P Kingma and Jimmy Ba. Adam: A method for stochastic optimization. arXiv:1412.6980, 2014

    2014 arXiv

  5. [13]

    Roarnet: A robust 3d object detection based on region approximation refinement

    Youngwook Paul Kwon Kiwoo Shin and Masayoshi Tomizuka. Roarnet: A robust 3d object detection based on region approximation refinement. arXiv:1811.03818, 2018

    2018 arXiv

  6. [14]

    Hypernet: Towards accurate region proposal generation and joint object detection

    Tao Kong, Anbang Yao, Yurong Chen, and Fuchun Sun. Hypernet: Towards accurate region proposal generation and joint object detection. In CVPR, 2016

    2016

  7. [15]

    Joint 3d proposal generation and ob- ject detection from view aggregation

    Jason Ku, Melissa Mozifian, Jungwook Lee, Ali Harakeh, and Steven Waslander. Joint 3d proposal generation and ob- ject detection from view aggregation. IROS, 2018

    2018

  8. [16]

    Pointpillars: Fast encoders for object detection from point clouds

    Alex H Lang, Sourabh V ora, Holger Caesar, Lubing Zhou, Jiong Yang, and Oscar Beijbom. Pointpillars: Fast encoders for object detection from point clouds. arXiv:1812.05784, 2018

    2018 arXiv

  9. [17]

    3d fully convolutional network for vehicle detection in point cloud

    Bo Li. 3d fully convolutional network for vehicle detection in point cloud. In IROS, 2017

    2017

  10. [18]

    Vehicle detection from 3d lidar using fully convolutional network

    Bo Li, Tianlei Zhang, and Tian Xia. Vehicle detection from 3d lidar using fully convolutional network. Robotics: Sci- ence and Systems, 2016

    2016

  11. [19]

    Pointcnn

    Yangyan Li, Rui Bu, Mingchao Sun, and Baoquan Chen. Pointcnn. NIPS, 2018

    2018

  12. [20]

    Deep continuous fusion for multi-sensor 3d object detection

    Ming Liang, Bin Yang, Shenlong Wang, and Raquel Urtasun. Deep continuous fusion for multi-sensor 3d object detection. In ECCV, 2018

    2018

  13. [21]

    Feature pyramid networks for object detection

    Tsung-Yi Lin, Piotr Doll ´ar, Ross B Girshick, Kaiming He, Bharath Hariharan, and Serge J Belongie. Feature pyramid networks for object detection. In CVPR, 2017

    2017

  14. [22]

    Ssd: Single shot multibox detector

    Wei Liu, Dragomir Anguelov, Dumitru Erhan, Christian Szegedy, Scott Reed, Cheng-Yang Fu, and Alexander C Berg. Ssd: Single shot multibox detector. In ECCV, 2016

    2016

  15. [23]

    Fast and furi- ous: Real time end-to-end 3d detection, tracking and motion forecasting with a single convolutional net

    Wenjie Luo, Bin Yang, and Raquel Urtasun. Fast and furi- ous: Real time end-to-end 3d detection, tracking and motion forecasting with a single convolutional net. In CVPR, 2018

    2018

  16. [24]

    V oxnet: A 3d con- volutional neural network for real-time object recognition

    Daniel Maturana and Sebastian Scherer. V oxnet: A 3d con- volutional neural network for real-time object recognition. In IROS, 2015

    2015

  17. [25]

    Qi, Wei Liu, Chenxia Wu, Hao Su, and Leonidas J

    Charles R. Qi, Wei Liu, Chenxia Wu, Hao Su, and Leonidas J. Guibas. Frustum pointnets for 3d object detec- tion from rgb-d data. In CVPR, 2018

    2018

  18. [26]

    Qi, Hao Su, Kaichun Mo, and Leonidas J Guibas

    Charles R. Qi, Hao Su, Kaichun Mo, and Leonidas J Guibas. Pointnet: Deep learning on point sets for 3d classification and segmentation. CVPR, 2017

    2017

  19. [27]

    Qi, Hao Su, Matthias Nießner, Angela Dai, Mengyuan Yan, and Leonidas J Guibas

    Charles R. Qi, Hao Su, Matthias Nießner, Angela Dai, Mengyuan Yan, and Leonidas J Guibas. V olumetric and multi-view cnns for object classification on 3d data. In CVPR, 2016

    2016

  20. [28]

    Qi, Li Yi, Hao Su, and Leonidas J Guibas

    Charles R. Qi, Li Yi, Hao Su, and Leonidas J Guibas. Point- net++: Deep hierarchical feature learning on point sets in a metric space. In NIPS, 2017

    2017

  21. [29]

    Pointr- cnn: 3d object proposal generation and detection from point cloud

    Shaoshuai Shi, Xiaogang Wang, and Hongsheng Li. Pointr- cnn: 3d object proposal generation and detection from point cloud. arXiv:1812.04244, 2018

    2018 arXiv

  22. [30]

    Training region-based object detectors with online hard ex- ample mining

    Abhinav Shrivastava, Abhinav Gupta, and Ross Girshick. Training region-based object detectors with online hard ex- ample mining. In CVPR, 2016

    2016

  23. [31]

    Complex-yolo: Real-time 3d object detec- tion on point clouds

    Martin Simon, Stefan Milz, Karl Amende, and Horst- Michael Gross. Complex-yolo: Real-time 3d object detec- tion on point clouds. arXiv:1803.06199, 2018

    2018 arXiv

  24. [32]

    Sliding shapes for 3d ob- ject detection in depth images

    Shuran Song and Jianxiong Xiao. Sliding shapes for 3d ob- ject detection in depth images. In ECCV, 2014

    2014

  25. [33]

    Deep sliding shapes for amodal 3d object detection in rgb-d images

    Shuran Song and Jianxiong Xiao. Deep sliding shapes for amodal 3d object detection in rgb-d images. In CVPR, 2016

    2016

  26. [34]

    Multi-view convolutional neural networks for 3d shape recognition

    Hang Su, Subhransu Maji, Evangelos Kalogerakis, and Erik Learned-Miller. Multi-view convolutional neural networks for 3d shape recognition. In ICCV, 2015

    2015

  27. [35]

    Multi-view 3d models from single images with a convolu- tional network

    Maxim Tatarchenko, Alexey Dosovitskiy, and Thomas Brox. Multi-view 3d models from single images with a convolu- tional network. In ECCV, 2016

    2016

  28. [36]

    Dynamic graph cnn for learning on point clouds

    Yue Wang, Yongbin Sun, Ziwei Liu, Sanjay E Sarma, Michael M Bronstein, and Justin M Solomon. Dynamic graph cnn for learning on point clouds. arXiv:1801.07829, 2018

    2018 arXiv

  29. [37]

    3d shapenets: A deep representation for volumetric shapes

    Zhirong Wu, Shuran Song, Aditya Khosla, Fisher Yu, Lin- guang Zhang, Xiaoou Tang, and Jianxiong Xiao. 3d shapenets: A deep representation for volumetric shapes. In CVPR, 2015

    2015

  30. [38]

    Pointfu- sion: Deep sensor fusion for 3d bounding box estimation

    Danfei Xu, Dragomir Anguelov, and Ashesh Jain. Pointfu- sion: Deep sensor fusion for 3d bounding box estimation. In CVPR, 2018

    2018

  31. [39]

    Second: Sparsely embed- ded convolutional detection

    Yan Yan, Yuxing Mao, and Bo Li. Second: Sparsely embed- ded convolutional detection. Sensors, 2018

    2018

  32. [40]

    Pixor: Real- time 3d object detection from point clouds

    Bin Yang, Wenjie Luo, and Raquel Urtasun. Pixor: Real- time 3d object detection from point clouds. In CVPR, 2018

    2018

  33. [41]

    Ipod: Intensive point-based object detector for point cloud

    Zetong Yang, Yanan Sun, Shu Liu, Xiaoyong Shen, and Ji- aya Jia. Ipod: Intensive point-based object detector for point cloud. arXiv:1812.05276, 2018

    2018 arXiv

  34. [42]

    mixup: Beyond empirical risk minimiza- tion

    Hongyi Zhang, Moustapha Cisse, Yann N Dauphin, and David Lopez-Paz. mixup: Beyond empirical risk minimiza- tion. In ICLR, 2017

    2017

  35. [43]

    V oxelnet: End-to-end learning for point cloud based 3d object detection

    Yin Zhou and Oncel Tuzel. V oxelnet: End-to-end learning for point cloud based 3d object detection. In CVPR, 2018

    2018

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.