Pith. sign in

REVIEW 4 major objections 4 minor 44 references

PromptDet: A Lightweight 3D Object Detection Framework with LiDAR Prompts

T0 review · 4 major / 4 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read PromptDet treats LiDAR as a prompt, adding fewer than 2% parameters to camera-only BEV detectors; it reports gains up to 22.8 mAP and 21.1 NDS in multi-modal mode and up to 2.4 mAP and 4.0 NDS in camera-only mode.

desk verdict Solid engineering result: a lightweight LiDAR fusion branch that also improves camera-only inference, but the camera-only gains are conditional on LiDAR at training time and the paper should say so more clearly. read the letter →

arxiv 2412.12460 v2 pith:IJ3OSRWS submitted 2024-12-17 cs.CV

Kun Guo , Qiang Ling This is my paper · ORCID
classification cs.CV
keywords 3Dobjectdetectionmulti-modalfusionpromptlearningbird's-eyeviewLiDAR-cameraknowledgedistillationnuScenesautonomousdriving
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

The paper tries to establish that LiDAR does not need a full fusion network or a pre-trained teacher to help a camera-based 3D detector. The missing depth geometry can be injected through a small 'prompter' made of two modules, AHA and CMKI, leaving the base detector nearly unchanged. If this is right, a single one-stage training run yields two usable detectors: a strong LiDAR-camera fusion detector and an improved camera-only detector that still beats the original camera baseline. The reported gains on nuScenes are up to 22.8% mAP and 21.1% NDS for fusion, and up to 2.4% mAP and 4.0% NDS for camera-only inference, with parameter growth below 2% and nearly unchanged camera inference time.

What carries the argument

Adaptive Hierarchical Aggregation (AHA) and Cross-Modal Knowledge Injection (CMKI) together form the LiDAR-assisted prompter. AHA computes softmax attention weights over pairs of LiDAR voxel features and camera pseudo-voxel features at three voxel scales, combines them, and flattens the result into a fusion BEV feature. CMKI first maps the camera BEV feature through a 3D-then-2D convolution imitation module so it can mimic the fusion feature, then applies three distillation losses (feature, relation, response) from detached fusion features to camera features; detaching ensures fusion training is not dragged down by the camera branch. A LiDAR switch at inference selects which features enter the shared BEV encoder and detection head.

What would settle it

Train PromptDet on a camera-only dataset (no LiDAR at training time) and check whether the camera branch still beats the baseline; if it does, the gains are not from cross-modal knowledge, and if it does not, the paper's camera-only claim is confirmed to require LiDAR supervision. Alternatively, replace the fusion features used in CMKI with a fixed random noise feature of the same shape and retrain; any remaining camera-only gain would imply the loss is a regularizer rather than a vehicle for LiDAR geometry.

Watch

Extended reading notes

Core claim

The central claim is that LiDAR geometry can be treated as a prompt rather than as a second stream: the LiDAR-assisted prompter fuses point-cloud voxel features with camera pseudo-voxel features at three scales using learned attention weights (AHA), then uses the fused features to teach the camera branch online through feature, relation, and response distillation (CMKI). By supervising both branches with the same ground truth and detaching the fusion branch from the distillation gradients, the same network learns to answer with and without LiDAR. On the nuScenes benchmark, the authors report that this recipe improves BEVDet-family baselines by up to 22.8 mAP and 21.1 NDS in multi-modal mode with fewer than 2% extra parameters, and by up to 2.4 mAP and 4.0 NDS in camera-only mode with negligible inference overhead.

Load-bearing premise

The camera-only improvements only appear when paired LiDAR and camera data are available during training; if LiDAR is absent from the training set, the framework gives the camera branch no extra knowledge to learn.

Editorial extensions

If this is right

  • Multi-modal 3D detection can be bolted onto an existing camera detector in one training stage, without pretraining a separate teacher and without a second heavy backbone.
  • The same checkpoint covers both fusion inference and camera-only inference, so LiDAR failure at deployment need not disable detection or require a fallback model.
  • The camera-only branch's improvement transfers learned geometry into features, so under the reported settings it costs almost no extra inference time.
  • Because the prompter is plug-and-play, it can be expected to extend to other BEVDet-series detectors and to related multi-camera perception heads such as occupancy prediction and BEV segmentation (the paper suggests this).

Reading between the lines

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

  • The method could likely be reused with other auxiliary depth signals (e.g., radar or pseudo-LiDAR) as the 'prompt' source, since the prompter only needs voxel features aligned to the camera BEV space.
  • A testable extension is whether the camera-only gain is a distillation of depth/location information or just a regularizing effect of multi-task training; comparing against a camera-only model trained with the same losses but a scrambled or random fusion branch would isolate the LiDAR-specific contribution.
  • Because the gains are measured on nuScenes, the method's practical value depends on how well the learned prompting transfers to new cities, camera rigs, or weather; a domain-shift benchmark would reveal how much of the improvement is dataset-specific.
  • The paper's parameter counts exclude LiDAR preprocessing, so the reported '+2%' is about network parameters only; an editorial reader should separate model size from total system cost when comparing with dedicated fusion detectors.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 4 minor

Summary. The paper proposes PromptDet, a plug-and-play LiDAR-assisted prompter for BEVDet-family camera-based 3D detectors. The method uses an Adaptive Hierarchical Aggregation (AHA) module to fuse LiDAR voxel features and camera pseudo-voxel features at three voxel scales, and a Cross-Modal Knowledge Injection (CMKI) module that distills knowledge from the detached fusion branch into the camera branch using UniDistill-style losses. The framework is trained in a single stage with a hybrid supervision strategy in which both branches share the BEV encoder and detection head. Experiments on nuScenes show substantial multi-modal gains (up to 22.8 mAP points and 21.1 NDS points with fewer than 2% extra parameters) and smaller camera-only inference gains (up to 2.4 mAP and 4.0 NDS), with the claim that the camera-only gains come with almost no impact on inference time.

Significance. The core contribution is a parameter-efficient fusion module that can be added to existing BEVDet-series detectors, with empirical validation across three baselines, component ablations, hyperparameter sensitivity, and efficiency comparisons. The released code and the reproducibility of the experiments are strengths. The paper's framing as 'prompt learning' is mostly analogical, but the technical content—online cross-modal distillation with gradient detachment—is sound and useful. The main value is practical: a user can obtain multi-modal fusion performance and an improved camera-only mode from a single-stage training run.

major comments (4)
  1. [Abstract; §Training and Inference] The camera-only improvement reported in the abstract ('Without LiDAR points, PromptDet still achieves an improvement...') is only available when LiDAR data were present during training, because CMKI requires a fusion branch and AHA requires LiDAR features. The methodology section states this dependency, but the abstract and conclusion do not; please scope the claim explicitly, for example by saying 'when LiDAR is used during training but is unavailable at inference, PromptDet-C remains improved.'
  2. [§Model Efficiency; Table 5] The claim that PromptDet-C has 'almost no impact on camera detection inference time' is not supported by any measurement. Table 5 reports latency only for the L&C mode of +LaP, not for the camera-only branch. Please measure and report the inference latency of the camera-only baseline with and without the imitation module, or qualify the claim as an estimate based on the two added convolution layers.
  3. [Abstract; Table 2] The increases of 22.8% mAP and 21.1% NDS are absolute percentage-point differences (e.g., 50.5 vs. 27.7 mAP in Table 2), not relative improvements. Please use 'percentage points' throughout the abstract, introduction, and main text to avoid misreading.
  4. [Introduction; §Model Efficiency] The paper motivates PromptDet by the training cost of KD-based methods, but no experiment measures training time, memory, or FLOPs against a KD baseline such as UniDistill or SimDistill. Since the 'single-stage, lightweight' claim is central to the positioning, please add such a comparison or temper the qualitative claims about training cost.
minor comments (4)
  1. [Methodology] The cross-references in the first paragraph of §Methodology are empty ('In Section , we introduce'). Please insert the correct section numbers.
  2. [Figure 2; §Training and Inference] The 'LiDAR modality switch' semantics are confusing: in §Training and Inference, the switch is 'turned off' for multi-modal detection and 'turned on' for camera-only detection, which inverts the usual meaning of a switch. Please rename it (e.g., 'LiDAR bypass switch') and describe its state unambiguously.
  3. [Figure 2 caption] There is a typo, 'PrompDet', in the caption of Figure 2.
  4. [Experimental Setup] The 'dynamic point cloud voxelization' is attributed to (Zhou et al. 2020), but that reference appears to describe an end-to-end multi-view fusion method rather than dynamic voxelization; please clarify the reference or cite the correct source.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: all reported gains are external benchmark measurements, with no equation-level reduction to inputs and no load-bearing self-citation.

full rationale

PromptDet's reported gains are empirical measurements on the nuScenes benchmark against fixed baselines. The multi-modal mAP/NDS improvements follow from training the AHA fusion module alongside a shared BEV encoder and detection head, while the camera-only improvements follow from the CMKI distillation losses defined in Eq. (7), which match camera features to detached fusion features. Neither quantity is defined in terms of the method's own output, and no fitted parameter is renamed as a prediction. The method borrows UniDistill's distillation losses from independent prior work (Zhou et al. 2023), and the AHA and imitation modules are newly proposed and ablated. The camera-only gains are conditional on LiDAR being available during training, since CMKI requires a fusion branch from which to distill, and AHA requires LiDAR points to produce fusion features. This is a deployment scoping condition, not a circular reduction: the abstract's phrase 'Without LiDAR points' refers to inference time, and the paper explicitly states both inference situations in the 'Training and Inference' section. No load-bearing self-citation, imported uniqueness theorem, or ansatz-smuggling citation appears; the authors' own prior work is not cited in the load-bearing argument. The paper is self-contained against an external benchmark, so the honest finding is no significant circularity.

Assumptions & free parameters 2 free parameters · 4 assumptions · 0 invented entities

No new physical or architectural entities are postulated; the additional components are standard convolutional and distillation modules. The main latent assumptions are the suitability of shared voxel fusion and shared-head training, plus the need for LiDAR during training.

free parameters (2)
  • CMKI loss weights λ1, λ2, λ3 = BEVDet: 1.1/8.0/2.0; BEVDet4D: 1.5/10.0/2.5; BEVDepth: 8.0/25.0/10.0
    Chosen per base model via sensitivity analysis (Table 6); they balance detection losses and the three distillation terms, and the results vary with them.
  • Moderate voxel size = [0.8m, 0.8m, 0.8m]
    Set by hand as the base voxel scale for fusion in AHA; affects the granularity of the hierarchical aggregation.
assumptions (4)
  • domain assumption nuScenes ground-truth annotations and official evaluation metrics are reliable.
    All conclusions are drawn from this single benchmark.
  • domain assumption Camera pseudo-voxel features and LiDAR voxel features can be aligned and fused in a shared voxel grid at multiple scales.
    Core to AHA; if the modalities cannot be meaningfully aligned, fusion gains would not materialize.
  • ad hoc to paper A single BEV encoder and detection head can serve both the fusion branch and the camera branch without performance collapse.
    The shared-head design is unusual and the paper's ablations show the need to detach fusion features for stability.
  • domain assumption UniDistill's feature, relation, and response distillation losses are suitable for online self-distillation between the fusion and camera branches.
    The CMKI losses are directly borrowed and adapted from UniDistill (cited).

how reviews work

0 comments
Cite this review

Pith. "Pith review of PromptDet: A Lightweight 3D Object Detection Framework with LiDAR Prompts." pith.science (2026). https://pith.science/paper/IJ3OSRWS

@misc{pith2026241212460,
  author       = {Pith},
  title        = {Pith review of: PromptDet: A Lightweight 3D Object Detection Framework with LiDAR Prompts},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/IJ3OSRWS}},
  note         = {Machine review of arXiv:2412.12460}
}
read the original abstract

Multi-camera 3D object detection aims to detect and localize objects in 3D space using multiple cameras, which has attracted more attention due to its cost-effectiveness trade-off. However, these methods often struggle with the lack of accurate depth estimation caused by the natural weakness of the camera in ranging. Recently, multi-modal fusion and knowledge distillation methods for 3D object detection have been proposed to solve this problem, which are time-consuming during the training phase and not friendly to memory cost. In light of this, we propose PromptDet, a lightweight yet effective 3D object detection framework motivated by the success of prompt learning in 2D foundation model. Our proposed framework, PromptDet, comprises two integral components: a general camera-based detection module, exemplified by models like BEVDet and BEVDepth, and a LiDAR-assisted prompter. The LiDAR-assisted prompter leverages the LiDAR points as a complementary signal, enriched with a minimal set of additional trainable parameters. Notably, our framework is flexible due to our prompt-like design, which can not only be used as a lightweight multi-modal fusion method but also as a camera-only method for 3D object detection during the inference phase. Extensive experiments on nuScenes validate the effectiveness of the proposed PromptDet. As a multi-modal detector, PromptDet improves the mAP and NDS by at most 22.8\% and 21.1\% with fewer than 2\% extra parameters compared with the camera-only baseline. Without LiDAR points, PromptDet still achieves an improvement of at most 2.4\% mAP and 4.0\% NDS with almost no impact on camera detection inference time.

Figures

Figures reproduced from arXiv: 2412.12460 by the authors.

Figure 1
Figure 1. Comparison of our PromptDet with previous de [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. The overview of our proposed PrompDet. The model is composed of a camera-only detector and the LiDAR-assisted [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Illustration of Adaptive Hierarchical Aggregation [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: Illustration of data distribution change during [PITH_FULL_IMAGE:figures/full_fig_p005_4.png]
Figure 5
Figure 5. Figure 5: Illustration to show the effect of Cross-Modal [PITH_FULL_IMAGE:figures/full_fig_p007_5.png]
Figure 6
Figure 6. Figure 6: The comparison of BEVDet (Huang et al. 2021) [PITH_FULL_IMAGE:figures/full_fig_p007_6.png]
Figure 7
Figure 7. Figure 7: Comparison of the baseline BEVDet and our proposed PromptDet. The red boxes and green boxes denote the ground [PITH_FULL_IMAGE:figures/full_fig_p009_7.png]
Figure 8
Figure 8. Figure 8: Visualization of feature maps in baseline and our proposed PromptDet. The red boxes denote the ground truth in the [PITH_FULL_IMAGE:figures/full_fig_p010_8.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

44 extracted references · 16 canonical work pages

  1. [1]

    Bahng, H.; Jahanian, A.; Sankaranarayanan, S.; and Isola, P. 2022. Exploring visual prompts for adapting large-scale models. arXiv preprint arXiv:2203.17274

    arXiv 2022

  2. [2]

    H.; Vora, S.; Liong, V

    Caesar, H.; Bankiti, V.; Lang, A. H.; Vora, S.; Liong, V. E.; Xu, Q.; Krishnan, A.; Pan, Y.; Baldan, G.; and Beijbom, O. 2020. nuscenes: A multimodal dataset for autonomous driving. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 11621--11631

    2020

  3. [3]

    Chen, S.; Ge, C.; Tong, Z.; Wang, J.; Song, Y.; Wang, J.; and Luo, P. 2022 a . Adaptformer: Adapting vision transformers for scalable visual recognition. Advances in Neural Information Processing Systems, 35: 16664--16678

    work page 2022

  4. [4]

    Chen, Z.; Li, Z.; Zhang, S.; Fang, L.; Jiang, Q.; and Zhao, F. 2022 b . Bevdistill: Cross-modal bev distillation for multi-view 3d object detection. arXiv preprint arXiv:2211.09386

    arXiv 2022

  5. [5]

    Contributors, M. 2020. MMDetection3D: OpenMMLab next-generation platform for general 3D object detection

    work page 2020

  6. [6]

    Deng, J.; Zhang, S.; Dayoub, F.; Ouyang, W.; Zhang, Y.; and Reid, I. 2024. PoIFusion: Multi-Modal 3D Object Detection via Fusion at Points of Interest. arXiv preprint arXiv:2403.09212

    arXiv 2024

  7. [7]

    Dosovitskiy, A.; Beyer, L.; Kolesnikov, A.; Weissenborn, D.; Zhai, X.; Unterthiner, T.; Dehghani, M.; Minderer, M.; Heigold, G.; Gelly, S.; et al. 2020. An image is worth 16x16 words: Transformers for image recognition at scale. arXiv preprint arXiv:2010.11929

    arXiv 2020

  8. [8]

    Ge, C.; Chen, J.; Xie, E.; Wang, Z.; Hong, L.; Lu, H.; Li, Z.; and Luo, P. 2023. Metabev: Solving sensor failures for 3d detection and map segmentation. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 8721--8731

    work page 2023

Show all 44 references
  1. [9]

    He, K.; Zhang, X.; Ren, S.; and Sun, J. 2016. Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition, 770--778

    2016

  2. [10]

    Hinton, G.; Vinyals, O.; and Dean, J. 2015. Distilling the knowledge in a neural network. arXiv preprint arXiv:1503.02531

    2015 arXiv

  3. [11]

    Huang, J.; and Huang, G. 2022. Bevdet4d: Exploit temporal cues in multi-camera 3d object detection. arXiv preprint arXiv:2203.17054

    2022 arXiv

  4. [12]

    Huang, J.; Huang, G.; Zhu, Z.; Ye, Y.; and Du, D. 2021. Bevdet: High-performance multi-camera 3d object detection in bird-eye-view. arXiv preprint arXiv:2112.11790

    2021 arXiv

  5. [13]

    Jiao, Y.; Jie, Z.; Chen, S.; Chen, J.; Ma, L.; and Jiang, Y.-G. 2023. Msmdfusion: Fusing lidar and camera at multiple scales with multi-depth seeds for 3d object detection. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 21643--21652

    2023

  6. [14]

    Jie, S.; and Deng, Z.-H. 2022. Convolutional bypasses are better vision transformer adapters. arXiv preprint arXiv:2207.07039

    2022 arXiv

  7. [15]

    H.; Vora, S.; Caesar, H.; Zhou, L.; Yang, J.; and Beijbom, O

    Lang, A. H.; Vora, S.; Caesar, H.; Zhou, L.; Yang, J.; and Beijbom, O. 2019. Pointpillars: Fast encoders for object detection from point clouds. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 12697--12705

    2019

  8. [16]

    Lester, B.; Al-Rfou, R.; and Constant, N. 2021. The power of scale for parameter-efficient prompt tuning. arXiv preprint arXiv:2104.08691

    2021 arXiv

  9. [17]

    Li, Y.; Chen, Y.; Qi, X.; Li, Z.; Sun, J.; and Jia, J. 2022 a . Unifying voxel-based representation with transformer for 3d object detection. Advances in Neural Information Processing Systems, 35: 18442--18455

    2022

  10. [18]

    Li, Y.; Ge, Z.; Yu, G.; Yang, J.; Wang, Z.; Shi, Y.; Sun, J.; and Li, Z. 2023 a . Bevdepth: Acquisition of reliable depth for multi-view 3d object detection. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 37, 1477--1485

    2023

  11. [19]

    Li, Z.; Wang, W.; Li, H.; Xie, E.; Sima, C.; Lu, T.; Qiao, Y.; and Dai, J. 2022 b . Bevformer: Learning bird’s-eye-view representation from multi-camera images via spatiotemporal transformers. In European conference on computer vision, 1--18. Springer

    2022

  12. [20]

    Li, Z.; Yu, Z.; Wang, W.; Anandkumar, A.; Lu, T.; and Alvarez, J. M. 2023 b . Fb-bev: Bev representation from forward-backward view transformations. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 6919--6928

    2023

  13. [21]

    Liang, T.; Xie, H.; Yu, K.; Xia, Z.; Lin, Z.; Wang, Y.; Tang, T.; Wang, B.; and Tang, Z. 2022. Bevfusion: A simple and robust lidar-camera fusion framework. Advances in Neural Information Processing Systems, 35: 10421--10434

    2022

  14. [22]

    Liu, P.; Yuan, W.; Fu, J.; Jiang, Z.; Hayashi, H.; and Neubig, G. 2023 a . Pre-train, prompt, and predict: A systematic survey of prompting methods in natural language processing. ACM Computing Surveys, 55(9): 1--35

    2023

  15. [23]

    Liu, Y.; Wang, T.; Zhang, X.; and Sun, J. 2022. Petr: Position embedding transformation for multi-view 3d object detection. In European Conference on Computer Vision, 531--548. Springer

    2022

  16. [24]

    L.; and Han, S

    Liu, Z.; Tang, H.; Amini, A.; Yang, X.; Mao, H.; Rus, D. L.; and Han, S. 2023 b . Bevfusion: Multi-task multi-sensor fusion with unified bird's-eye view representation. In 2023 IEEE international conference on robotics and automation (ICRA), 2774--2781. IEEE

    2023

  17. [25]

    Loshchilov, I.; and Hutter, F. 2017. Decoupled weight decay regularization. arXiv preprint arXiv:1711.05101

    2017 arXiv

  18. [26]

    Philion, J.; and Fidler, S. 2020. Lift, splat, shoot: Encoding images from arbitrary camera rigs by implicitly unprojecting to 3d. In Computer Vision--ECCV 2020: 16th European Conference, Glasgow, UK, August 23--28, 2020, Proceedings, Part XIV 16, 194--210. Springer

    2020

  19. [27]

    R.; Su, H.; Mo, K.; and Guibas, L

    Qi, C. R.; Su, H.; Mo, K.; and Guibas, L. J. 2017. Pointnet: Deep learning on point sets for 3d classification and segmentation. In Proceedings of the IEEE conference on computer vision and pattern recognition, 652--660

    2017

  20. [28]

    W.; Hallacy, C.; Ramesh, A.; Goh, G.; Agarwal, S.; Sastry, G.; Askell, A.; Mishkin, P.; Clark, J.; et al

    Radford, A.; Kim, J. W.; Hallacy, C.; Ramesh, A.; Goh, G.; Agarwal, S.; Sastry, G.; Askell, A.; Mishkin, P.; Clark, J.; et al. 2021. Learning transferable visual models from natural language supervision. In International conference on machine learning, 8748--8763. PMLR

    2021

  21. [29]

    Reading, C.; Harakeh, A.; Chae, J.; and Waslander, S. L. 2021. Categorical depth distribution network for monocular 3d object detection. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 8555--8564

    2021

  22. [30]

    Wang, S.; Caesar, H.; Nan, L.; and Kooij, J. F. 2024. Unibev: Multi-modal 3d object detection with uniform bev encoders for robustness against missing sensor modalities. In 2024 IEEE Intelligent Vehicles Symposium (IV), 2776--2783. IEEE

    2024

  23. [31]

    Wang, T.; Zhu, X.; Pang, J.; and Lin, D. 2021. Fcos3d: Fully convolutional one-stage monocular 3d object detection. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 913--922

    2021

  24. [32]

    C.; Zhang, T.; Wang, Y.; Zhao, H.; and Solomon, J

    Wang, Y.; Guizilini, V. C.; Zhang, T.; Wang, Y.; Zhao, H.; and Solomon, J. 2022. Detr3d: 3d object detection from multi-view images via 3d-to-2d queries. In Conference on Robot Learning, 180--191. PMLR

    2022

  25. [33]

    Wang, Z.; Li, D.; Luo, C.; Xie, C.; and Yang, X. 2023. Distillbev: Boosting multi-camera 3d object detection with cross-modal knowledge distillation. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 8637--8646

    2023

  26. [34]

    Yan, J.; Liu, Y.; Sun, J.; Jia, F.; Li, S.; Wang, T.; and Zhang, X. 2023. Cross modal transformer: Towards fast and robust 3d object detection. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 18268--18278

    2023

  27. [35]

    Yan, Y.; Mao, Y.; and Li, B. 2018. Second: Sparsely embedded convolutional detection. Sensors, 18(10): 3337

    2018

  28. [36]

    Zhang, Y.; Zhang, Q.; Hou, J.; Yuan, Y.; and Xing, G. 2023. Unleash the potential of image branch for cross-modal 3d object detection. In Advances in Neural Information Processing Systems, volume 36, 51562--51583

    2023

  29. [37]

    Zhao, H.; Zhang, Q.; Zhao, S.; Chen, Z.; Zhang, J.; and Tao, D. 2024. Simdistill: Simulated multi-modal distillation for bev 3d object detection. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 38, 7460--7468

    2024

  30. [38]

    Zhou, S.; Liu, W.; Hu, C.; Zhou, S.; and Ma, C. 2023. UniDistill: A Universal Cross-Modality Knowledge Distillation Framework for 3D Object Detection in Bird's-Eye View. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 5116--5125

    2023

  31. [39]

    Zhou, Y.; Sun, P.; Zhang, Y.; Anguelov, D.; Gao, J.; Ouyang, T.; Guo, J.; Ngiam, J.; and Vasudevan, V. 2020. End-to-end multi-view fusion for 3d object detection in lidar point clouds. In Conference on Robot Learning, 923--932. PMLR

    2020

  32. [40]

    Zhou, Y.; and Tuzel, O. 2018. Voxelnet: End-to-end learning for point cloud based 3d object detection. In Proceedings of the IEEE conference on computer vision and pattern recognition, 4490--4499

    2018

  33. [41]

    Zhu, B.; Jiang, Z.; Zhou, X.; Li, Z.; and Yu, G. 2019. Class-balanced grouping and sampling for point cloud 3d object detection. arXiv preprint arXiv:1908.09492

    2019 arXiv

  34. [42]

    Zhu, J.; Lai, S.; Chen, X.; Wang, D.; and Lu, H. 2023. Visual prompt multi-modal tracking. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 9516--9526

    2023

  35. [43]

    , " * write output.state after.block = add.period write newline

    ENTRY address archivePrefix author booktitle chapter edition editor eid eprint howpublished institution isbn journal key month note number organization pages publisher school series title type volume year label extra.label sort.label short.list INTEGERS output.state before.all...

  36. [44]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 gl...

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.