Pith. sign in

REVIEW 4 major objections 4 minor 45 references

HSDA: High-frequency Shuffle Data Augmentation for Bird's-Eye-View Map Segmentation

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 High-frequency pixel shuffling improves BEV map segmentation across networks and sets a new nuScenes state of the art.

desk verdict A cheap, model-agnostic augmentation that likely works as claimed, but the paper as written does not formally define the shuffle operation (Hermitian symmetry break), and the evaluation needs seeds and a cleaner SOTA comparison. read the letter →

arxiv 2412.06127 v1 pith:T7MWPNII submitted 2024-12-09 cs.CV

Calvin Glisson , Qiuxiao Chen This is my paper · ORCID
classification cs.CV
keywords dataaugmentationfrequencydomainFastFourierTransformBEVmapsegmentationautonomousdrivingnuSceneshigh-frequencyinformationGaussianfilter
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 argues that the high-frequency content of camera images carries the boundary information that bird's-eye-view (BEV) map segmentation most depends on, and that deliberately perturbing that content during training makes networks more robust. To exploit this, the authors propose High-frequency Shuffle Data Augmentation (HSDA): take one color channel from a multi-view camera image, transform it with the Fast Fourier Transform, split the spectrum with Gaussian filters, randomly shuffle the K strongest high-frequency coefficients, and invert the transform to obtain an augmented image that keeps the same BEV ground truth. Applied to several BEV segmentation models on nuScenes, the augmented training set improves mean IoU by 1.6 to 2.2 points, and combined with the RGC network it reaches 61.3% mIoU, surpassing published camera-only results by at least 1.6 points. The method also transfers to monocular 3D object detection, raising pedestrian and car AP on KITTI, which suggests a general role for frequency-domain augmentation in camera-based driving perception.

What carries the argument

The load-bearing object is the high-frequency shuffle operation. An input image's selected channel is sent through the Fast Fourier Transform; a Gaussian low-pass filter $G_L$ with width $D$ isolates the low-frequency spectrum, and the complement $G_H = 1 - G_L$ isolates high frequencies. The $K$ largest-magnitude coefficients in the high-frequency spectrum are randomly permuted, producing an augmented high-frequency spectrum; adding back the unmodified low-frequency spectrum and applying the inverse FFT yields an augmented channel. Because the low-frequency structure is untouched, the augmented image is assumed to share the same BEV segmentation ground truth as the original, while the shuffled high frequencies force the network to learn which high-frequency patterns are meaningful.

What would settle it

A direct test is to check whether the augmented image still aligns with the BEV ground truth at the pixel level: project ground-truth map boundaries back into the camera view, apply HSDA, and measure how far the shuffled high-frequency coefficients move those boundaries. If the shuffle shifts edge locations by more than a pixel for a noticeable fraction of images, the 'same ground truth' assumption is violated. Alternatively, compare HSDA to a control where the same high-frequency coefficients are replaced by random noise; if random noise performs equally well, the shuffle's specific permutation is irrelevant and the proposed explanation is unsupported.

Watch

Extended reading notes

Core claim

The central discovery is that a simple frequency-domain perturbation—shuffling the top K pixels by magnitude in the high-frequency spectrum of a single randomly chosen RGB channel—acts as an effective regularizer for BEV map segmentation. The paper shows experimentally that models trained with HSDA outperform the same models without it across three architectures (BEVFusion, a BEVDet-based baseline, and RGC), with no class regressing. The largest gain is in fine-grained categories such as stop lines and dividers, and the final RGC + HSDA model reports 61.3% mIoU on the nuScenes validation set, which the authors state is a new state-of-the-art for camera-only BEV map segmentation.

Load-bearing premise

The load-bearing premise is that randomly shuffling the strongest high-frequency pixels in one color channel leaves the BEV segmentation ground truth valid for the augmented image; if the shuffled pixels carry boundary structure the network relies on, the augmentation could teach it wrong labels.

Editorial extensions

If this is right

  • HSDA is model-agnostic: applying it to BEVFusion, the BEVDet-based baseline, and RGC improves mIoU by 2.2, 1.6, and 1.6 points respectively on nuScenes, with no class regressing.
  • Since HSDA requires no changes to network architecture or extra parameters, any existing BEV segmentation pipeline can adopt it as a drop-in augmentation during training.
  • The frequency-domain perspective suggests that controlling spectral content, rather than only spatial transforms such as flipping and scaling, is a productive direction for camera-based driving perception.
  • The improvement is concentrated in fine-grained classes such as stop lines, dividers, and pedestrian crossings, implying better edge and small-region perception.
  • The method transfers to monocular 3D object detection, improving mean AP on KITTI, so its benefit may extend beyond BEV segmentation to other camera-based driving tasks.

Reading between the lines

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

  • If the 'same ground truth' assumption holds, HSDA can be viewed as a consistency regularizer: the network must produce the same BEV map despite shuffled high-frequency content, a form of invariance training that could be combined with explicit consistency losses for further gains.
  • The success of shuffling high-frequency coefficients raises the question of whether other spectral perturbations, such as phase scrambling or selective band suppression, could provide complementary regularization; FDA is one such variant but underperforms HSDA in the reported experiments.
  • A natural stress test is to measure HSDA's effect under heavy sensor noise or rain, where the high-frequency band is already corrupted; the paper's qualitative rainy-scene result hints at a denoising benefit, but a quantitative study at varying noise levels would clarify the boundary of its advantage.
  • Because only one color channel is perturbed, the network may learn to rely on inter-channel consistency; testing HSDA on grayscale or hyperspectral inputs could reveal how much of the gain depends on RGB chroma.
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

4 major / 4 minor

Summary. The paper proposes HSDA, a frequency-domain data augmentation for camera-based BEV map segmentation. The method applies FFT to a randomly chosen RGB channel, separates low- and high-frequency components via Gaussian filters, randomly shuffles the top K high-frequency coefficients, recombines them with the low-frequency component, and inverse-transforms to produce an augmented image that is paired with the original BEV ground truth. Experiments on the nuScenes validation set across BEVFusion, a BEVDet-based baseline, and the authors' RGC network report consistent mIoU improvements of 1.6-2.2 points, with RGC+HSDA reaching 61.3 mIoU and claimed as camera-only state of the art. A KITTI MonoCon experiment additionally reports improved mean 3D AP for monocular detection.

Significance. HSDA is attractive in principle: it is model-agnostic, requires no architectural changes or extra parameters, is simple to implement, and the reported gains are internally consistent across three architectures and several semantic classes. The released code would allow others to verify the results. However, the method as described in Sec. 3.2 is not mathematically well-defined because the shuffle can destroy Hermitian symmetry, and the empirical SOTA claim rests on a validation-set comparison with hyperparameters selected on the same set and on one acknowledged mismatched temporal baseline. These issues must be resolved before the contribution can be properly assessed.

major comments (4)
  1. [Sec. 3.2, Eqs. (1)-(6)] The shuffle operation is not defined on conjugate-symmetric pairs. For a real input image, the FFT spectrum satisfies Hermitian symmetry, and the Gaussian filters are real and symmetric, so the low- and high-frequency components retain that symmetry. Randomly permuting the complex coefficients at the top-K magnitude locations almost surely breaks the pairing between (u,v) and (-u,-v), so the shuffled high-frequency spectrum is not Hermitian and the inverse FFT yields a complex-valued image. The text never states that the real part is taken, that rfft/irfft is used, or that coefficients are shuffled in conjugate-symmetric pairs. Since the augmented image is the central object of the paper, this is a load-bearing gap: as written, the augmented channel is undefined and the claim that the augmented image shares the same BEV segmentation ground truth is unsubstantiated. Please specify the exact inverse-transform operation, verify that the resulting image is real, and describe any clipping, rounding, or normalization applied.
  2. [Sec. 4.2, Tables 2-4 and 6] All experiments are reported on the nuScenes validation set, and the K and D hyperparameters are selected on that same set from three-point sweeps (Tables 2 and 3). The headline SOTA claim (Table 6) is then measured on the same validation set. With no error bars, multiple seeds, or a held-out split, it is unclear whether the 1.6-2.2 mIoU gains and the 1.6% margin to the next method are within run-to-run variation. Please report variance over at least three seeds or provide test-set results, and ideally fix K and D on a held-out portion before evaluating the headline configuration.
  3. [Sec. 4.2.2, Table 7] The comparison with PETRv2 is not a like-for-like SOTA comparison: PETRv2 uses temporal information while RGC+HSDA is single-frame, a mismatch the authors acknowledge in the text. The divider gain and drivable-area loss relative to PETRv2 are therefore not attributable to the augmentation alone. Please retrain or quote a single-frame PETRv2 variant, or explicitly remove PETRv2 from the SOTA claim and present Table 7 only as a reference point.
  4. [Sec. 3.2 and Fig. 3] The premise that shuffling high-frequency coefficients preserves the validity of the BEV segmentation label is asserted but not analyzed. If the shuffled coefficients are concentrated at genuine boundary structures, the inverse-transformed image can contain displaced edge or color artifacts that the network may learn to associate with incorrect labels. The grid-like artifacts and color-tinting effects described in Sec. 3.2 suggest this is a real risk. Please provide an analysis or experiment showing the label-preservation assumption holds, for example by comparing training with HSDA against a control augmentation that shuffles low-frequency coefficients, or by measuring segmentation accuracy under the real-part extraction choice.
minor comments (4)
  1. [Sec. 4.1] The dataset description says "1,400,00 camera images"; this should read "1,400,000".
  2. [Sec. 3.2] The notation "Shuf f le" contains an extra space, and the operation is described only as "a randomized swap for each pixel within the pool of the top K frequencies." Please clarify whether this is a random permutation, repeated pairwise swaps, or something else, and state whether the shuffle is applied independently per training sample and per epoch.
  3. [Sec. 4.2.2, Table 7] The source and configuration of the "single-timestamp BEVFormer" results are not given; please cite the specific checkpoint, training protocol, or paper section so the numbers can be reproduced.
  4. [Sec. 4.2.3, Table 8] The text says HSDA improves the mean AP overall, but the cyclist category declines at all difficulty levels; please discuss this trade-off explicitly in the text.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: HSDA's reported gains are empirical measurements, and the RGC self-citation and validation-set hyperparameter tuning do not make the improvement tautological.

full rationale

The paper does not claim a derivation; it reports an empirical augmentation study. The central claim (RGC+HSDA reaches 61.3 mIoU) is a measured outcome on the nuScenes validation set, not a consequence of Eqs. (1)-(6). The only candidates for circularity are (i) the SOTA baseline RGC is the authors' prior work [7], and (ii) K and D are selected on the same validation set used for the headline number. Neither reduces the claimed improvement to an input: RGC is retrained and compared against seven external SOTA methods in Table 6, and hyperparameter choice from a validation sweep is standard selection, not a fitted parameter disguised as a prediction. The assumption that I_Aug shares the same map segmentation ground truth is asserted rather than derived; if the shuffle breaks Hermitian symmetry, the inverse FFT may not yield a real image (Sec. 3.2, Eqs. (1)-(6)), but that is a formal correctness gap, not a circular step. No equation in the paper is equivalent by construction to the reported mIoU gain.

Assumptions & free parameters 2 free parameters · 3 assumptions · 0 invented entities

The headline result depends on two hand-tuned hyperparameters and on the unvalidated assumption that the shuffle keeps semantic content intact. No new physical or conceptual entities are introduced.

free parameters (2)
  • K = 2000
    Number of dominant high-frequency pixels shuffled; chosen from {1000, 2000, 3000} based on validation mIoU (Table 2).
  • D = 10
    Gaussian filter cutoff parameter in Eq. (4); chosen from {5, 10, 15} based on validation mIoU (Table 3).
assumptions (3)
  • standard math FFT/iFFT pair and Gaussian low-pass/high-pass decomposition reconstruct the image exactly when recombined (Eqs. 1 to 6).
    Relies on standard Fourier properties; the paper does not justify the specific filter shape beyond a standard Gaussian.
  • domain assumption Shuffling the top K high-frequency pixels of one color channel preserves the validity of the BEV segmentation ground truth for the augmented image.
    Stated in Sec. 3.2: the augmented image 'shares the same map segmentation ground truth as the input image'; this is the load-bearing label-preservation assumption.
  • domain assumption The network's BEV segmentation performance depends primarily on high-frequency image content, so perturbing high frequencies acts as beneficial regularization.
    Motivated by Table 1 high-frequency-only versus low-frequency-only baseline experiments, but the transfer to the full network is assumed.

how reviews work

0 comments
Cite this review

Pith. "Pith review of HSDA: High-frequency Shuffle Data Augmentation for Bird's-Eye-View Map Segmentation." pith.science (2026). https://pith.science/paper/T7MWPNII

@misc{pith2026241206127,
  author       = {Pith},
  title        = {Pith review of: HSDA: High-frequency Shuffle Data Augmentation for Bird's-Eye-View Map Segmentation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/T7MWPNII}},
  note         = {Machine review of arXiv:2412.06127}
}
read the original abstract

Autonomous driving has garnered significant attention in recent research, and Bird's-Eye-View (BEV) map segmentation plays a vital role in the field, providing the basis for safe and reliable operation. While data augmentation is a commonly used technique for improving BEV map segmentation networks, existing approaches predominantly focus on manipulating spatial domain representations. In this work, we investigate the potential of frequency domain data augmentation for camera-based BEV map segmentation. We observe that high-frequency information in camera images is particularly crucial for accurate segmentation. Based on this insight, we propose High-frequency Shuffle Data Augmentation (HSDA), a novel data augmentation strategy that enhances a network's ability to interpret high-frequency image content. This approach encourages the network to distinguish relevant high-frequency information from noise, leading to improved segmentation results for small and intricate image regions, as well as sharper edge and detail perception. Evaluated on the nuScenes dataset, our method demonstrates broad applicability across various BEV map segmentation networks, achieving a new state-of-the-art mean Intersection over Union (mIoU) of 61.3% for camera-only systems. This significant improvement underscores the potential of frequency domain data augmentation for advancing the field of autonomous driving perception. Code has been released: https://github.com/Zarhult/HSDA

Figures

Figures reproduced from arXiv: 2412.06127 by the authors.

Figure 1
Figure 1. Illustration of high-frequency spectrum, low-frequency spectrum and the corresponding images. (a) Original image. (b) High [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Overview of our baseline network architecture. It begins by processing multi-view camera images using an image encoder [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. The proposed High-frequency Shuffle Data Augmentation (HSDA) method introduces perturbations in the high-frequency [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Illustration of one sample rainy scene. Red circles highlight [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]
Figure 5
Figure 5. Figure 5: Qualitative results are presented for daytime, rainy, and nighttime scenarios. The left panels display multi-view input images, [PITH_FULL_IMAGE:figures/full_fig_p008_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

45 extracted references · 33 canonical work pages

  1. [1]

    Rs-aug: Improve 3d object de- tection on lidar with realistic simulator based data augmen- tation

    Pei An, Junxiong Liang, Jie Ma, Yanfei Chen, Liheng Wang, You Yang, and Qiong Liu. Rs-aug: Improve 3d object de- tection on lidar with realistic simulator based data augmen- tation. IEEE Transactions on Intelligent Transportation Sys- tems, 24(9):10165–10176, 2023. 3

    work page 2023

  2. [2]

    HS3: Learning with Proper Task Complexity in Hierarchically Supervised Semantic Segmentation

    Shubhankar Borse, Hong Cai, Yizhe Zhang, and Fatih Porikli. Hs3: Learning with proper task complexity in hier- archically supervised semantic segmentation. arXiv preprint arXiv:2111.02333, 2021. 2

    work page Pith review arXiv 2021

  3. [3]

    X-align: Cross-modal cross-view alignment for bird’s-eye-view segmentation

    Shubhankar Borse, Marvin Klingner, Varun Ravi Kumar, Hong Cai, Abdulaziz Almuzairee, Senthil Yogamani, and Fatih Porikli. X-align: Cross-modal cross-view alignment for bird’s-eye-view segmentation. In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, pages 3287–3297, 2023. 2, 5, 7

    work page 2023

  4. [4]

    Panoptic, instance and semantic relations: A relational context encoder to enhance panoptic segmentation

    Shubhankar Borse, Hyojin Park, Hong Cai, Debasmit Das, Risheek Garrepalli, and Fatih Porikli. Panoptic, instance and semantic relations: A relational context encoder to enhance panoptic segmentation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 1269–1279, 2022. 2

    work page 2022

  5. [5]

    Inverseform: A loss function for structured boundary-aware segmentation

    Shubhankar Borse, Ying Wang, Yizhe Zhang, and Fatih Porikli. Inverseform: A loss function for structured boundary-aware segmentation. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 5901–5911, 2021. 2

    work page 2021

  6. [6]

    nuscenes: A multi- modal dataset for autonomous driving

    Holger Caesar, Varun Bankiti, Alex H Lang, Sourabh V ora, Venice Erin Liong, Qiang Xu, Anush Krishnan, Yu Pan, Gi- ancarlo Baldan, and Oscar Beijbom. nuscenes: A multi- modal dataset for autonomous driving. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 11621–11631, 2020. 5

    2020

  7. [7]

    Residual graph convolutional network for bird’s-eye-view semantic segmentation

    Qiuxiao Chen and Xiaojun Qi. Residual graph convolutional network for bird’s-eye-view semantic segmentation. In Pro- ceedings of the IEEE/CVF Winter Conference on Applica- tions of Computer Vision, pages 3324–3331, 2024. 1, 2, 5, 7

    work page 2024

  8. [8]

    Bevseg: Geometry and data-driven based multi-view segmentation in bird’s-eye-view

    Qiuxiao Chen, Hung-Shuo Tai, Pengfei Li, Ke Wang, and Xiaojun Qi. Bevseg: Geometry and data-driven based multi-view segmentation in bird’s-eye-view. InInternational Conference on Computer Vision Systems , pages 432–443. Springer, 2023. 1, 2

    work page 2023

Show all 45 references
  1. [9]

    Monocular vehicle 3d bounding box estimation using homograhy and geome- try in traffic scene

    Yiqiang Chen, Feng Liu, and Ke Pei. Monocular vehicle 3d bounding box estimation using homograhy and geome- try in traffic scene. In ICASSP 2022-2022 IEEE Interna- tional Conference on Acoustics, Speech and Signal Process- ing (ICASSP), pages 1995–1999. IEEE, 2022. 2

    2022

  2. [10]

    Neat: Neural attention fields for end-to-end autonomous driving

    Kashyap Chitta, Aditya Prakash, and Andreas Geiger. Neat: Neural attention fields for end-to-end autonomous driving. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 15793–15803, 2021. 1

    2021

  3. [11]

    Real-time high-performance semantic image segmentation of urban street scenes

    Genshun Dong, Yan Yan, Chunhua Shen, and Hanzi Wang. Real-time high-performance semantic image segmentation of urban street scenes. IEEE Transactions on Intelligent Transportation Systems, 22(6):3258–3274, 2020. 2

    2020

  4. [12]

    Tbp-former: Learning temporal bird’s-eye-view pyramid for joint perception and prediction in vision-centric autonomous driving

    Shaoheng Fang, Zi Wang, Yiqi Zhong, Junhao Ge, and Si- heng Chen. Tbp-former: Learning temporal bird’s-eye-view pyramid for joint perception and prediction in vision-centric autonomous driving. In Proceedings of the IEEE/CVF Con- ference on Computer Vision and Pattern Recognit...

    2023

  5. [13]

    3d-lanenet: end-to-end 3d multiple lane detection

    Noa Garnett, Rafi Cohen, Tomer Pe’er, Roee Lahav, and Dan Levi. 3d-lanenet: end-to-end 3d multiple lane detection. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 2921–2930, 2019. 2

    2019

  6. [14]

    Metabev: Solving sensor failures for 3d detection and map segmentation

    Chongjian Ge, Junsong Chen, Enze Xie, Zhongdao Wang, Lanqing Hong, Huchuan Lu, Zhenguo Li, and Ping Luo. Metabev: Solving sensor failures for 3d detection and map segmentation. In Proceedings of the IEEE/CVF Interna- tional Conference on Computer Vision , pages 8721–8731,

  7. [15]

    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 Conference on Computer Vision and Pattern Recog- nition (CVPR), 2012. 7

    2012

  8. [16]

    Frequency-enhanced data augmentation for vision-and-language navigation

    Keji He, Chenyang Si, Zhihe Lu, Yan Huang, Liang Wang, and Xinchao Wang. Frequency-enhanced data augmentation for vision-and-language navigation. Advances in Neural In- formation Processing Systems, 36, 2024. 3, 6

    2024

  9. [17]

    Fiery: Future instance prediction in bird’s- eye view from surround monocular cameras

    Anthony Hu, Zak Murez, Nikhil Mohan, Sof ´ıa Dudas, Jef- frey Hawke, Vijay Badrinarayanan, Roberto Cipolla, and Alex Kendall. Fiery: Future instance prediction in bird’s- eye view from surround monocular cameras. In Proceedings of the IEEE/CVF International Conference on Compu...

    2021

  10. [18]

    Learning im- plicit feature alignment function for semantic segmentation

    Hanzhe Hu, Yinbo Chen, Jiarui Xu, Shubhankar Borse, Hong Cai, Fatih Porikli, and Xiaolong Wang. Learning im- plicit feature alignment function for semantic segmentation. In European Conference on Computer Vision , pages 487–

  11. [19]

    Pattern-aware data augmentation for lidar 3d object detection

    Jordan SK Hu and Steven L Waslander. Pattern-aware data augmentation for lidar 3d object detection. In 2021 IEEE International Intelligent Transportation Systems Conference (ITSC), pages 2703–2710. IEEE, 2021. 3

    2021

  12. [20]

    Planning-oriented autonomous driving

    Yihan Hu, Jiazhi Yang, Li Chen, Keyu Li, Chonghao Sima, Xizhou Zhu, Siqi Chai, Senyao Du, Tianwei Lin, Wenhai Wang, et al. Planning-oriented autonomous driving. In Pro- ceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 17853–17862, 2023. 1

    2023

  13. [21]

    Bevdet: High-performance multi-camera 3d object de- tection in bird-eye-view

    Junjie Huang, Guan Huang, Zheng Zhu, Yun Ye, and Dalong Du. Bevdet: High-performance multi-camera 3d object de- tection in bird-eye-view. arXiv preprint arXiv:2112.11790,

    arXiv

  14. [22]

    Data augmen- tation techniques in time series domain: a survey and tax- onomy

    Guillermo Iglesias, Edgar Talavera, ´Angel Gonz´alez-Prieto, Alberto Mozo, and Sandra G ´omez-Canaval. Data augmen- tation techniques in time series domain: a survey and tax- onomy. Neural Computing and Applications, 35(14):10123– 10145, 2023. 3

    2023

  15. [23]

    Ddp: Diffusion model for dense visual prediction

    Yuanfeng Ji, Zhe Chen, Enze Xie, Lanqing Hong, Xihui Liu, Zhaoqiang Liu, Tong Lu, Zhenguo Li, and Ping Luo. Ddp: Diffusion model for dense visual prediction. In Proceedings of the IEEE/CVF International Conference on Computer Vi- sion, pages 21741–21752, 2023. 1, 2, 5, 7

    2023

  16. [24]

    En- hancing monocular 3d object detection through data aug- mentation strategies

    Yisong Jia, Jue Wang, Huihui Pan, and Weichao Sun. En- hancing monocular 3d object detection through data aug- mentation strategies. IEEE Transactions on Instrumentation and Measurement, 2024. 3

    2024

  17. [25]

    Bevdepth: Acquisition of reliable depth for multi-view 3d object detec- tion

    Yinhao Li, Zheng Ge, Guanyi Yu, Jinrong Yang, Zengran Wang, Yukang Shi, Jianjian Sun, and Zeming Li. Bevdepth: Acquisition of reliable depth for multi-view 3d object detec- tion. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 37, pages 1477–1485, 2023. 3

    2023

  18. [26]

    Bevformer: Learning bird’s-eye-view representation from multi-camera images via spatiotemporal transformers

    Zhiqi Li, Wenhai Wang, Hongyang Li, Enze Xie, Chong- hao Sima, Tong Lu, Qiao Yu, and Jifeng Dai. Bevformer: Learning bird’s-eye-view representation from multi-camera images via spatiotemporal transformers. arXiv preprint arXiv:2203.17270, 2022. 1, 2, 7

    2022 arXiv

  19. [27]

    Timesurl: Self-supervised contrastive learning for universal time series representation learning

    Jiexi Liu and Songcan Chen. Timesurl: Self-supervised contrastive learning for universal time series representation learning. In Proceedings of the AAAI Conference on Artifi- cial Intelligence, volume 38, pages 13918–13926, 2024. 3

    2024

  20. [28]

    Bird’s-eye-view scene graph for vision-language navigation

    Rui Liu, Xiaohan Wang, Wenguan Wang, and Yi Yang. Bird’s-eye-view scene graph for vision-language navigation. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 10968–10980, 2023. 1

    2023

  21. [29]

    Learning auxil- iary monocular contexts helps monocular 3d object detec- tion

    Xianpeng Liu, Nan Xue, and Tianfu Wu. Learning auxil- iary monocular contexts helps monocular 3d object detec- tion. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 36, pages 1810–1818, 2022. 7

    2022

  22. [30]

    Petrv2: A unified framework for 3d perception from multi-camera images

    Yingfei Liu, Junjie Yan, Fan Jia, Shuailin Li, Aqi Gao, Tian- cai Wang, and Xiangyu Zhang. Petrv2: A unified framework for 3d perception from multi-camera images. InProceedings of the IEEE/CVF International Conference on Computer Vi- sion, pages 3262–3272, 2023. 2, 7

    2023

  23. [31]

    Bevfusion: Multi- task multi-sensor fusion with unified bird’s-eye view repre- sentation

    Zhijian Liu, Haotian Tang, Alexander Amini, Xinyu Yang, Huizi Mao, Daniela L Rus, and Song Han. Bevfusion: Multi- task multi-sensor fusion with unified bird’s-eye view repre- sentation. In 2023 IEEE international conference on robotics and automation (ICRA) , pages 2774–2781. ...

    2023

  24. [32]

    Decoupled weight decay regularization

    Ilya Loshchilov and Frank Hutter. Decoupled weight decay regularization. arXiv preprint arXiv:1711.05101, 2017. 5

    2017 arXiv

  25. [33]

    Driving among flatmobiles: Bird-eye-view occu- pancy grids from a monocular camera for holistic trajectory planning

    Abdelhak Loukkal, Yves Grandvalet, Tom Drummond, and You Li. Driving among flatmobiles: Bird-eye-view occu- pancy grids from a monocular camera for holistic trajectory planning. In Proceedings of the IEEE/CVF Winter Con- ference on Applications of Computer Vision , pages 51–60,

  26. [34]

    Lift, splat, shoot: Encod- ing images from arbitrary camera rigs by implicitly unpro- jecting to 3d

    Jonah Philion and Sanja Fidler. Lift, splat, shoot: Encod- ing images from arbitrary camera rigs by implicitly unpro- jecting to 3d. In Computer Vision–ECCV 2020: 16th Euro- pean Conference, Glasgow, UK, August 23–28, 2020, Pro- ceedings, Part XIV 16, pages 194–210. Springer, ...

    2020

  27. [35]

    Ortho- graphic feature transform for monocular 3d object detection

    Thomas Roddick, Alex Kendall, and Roberto Cipolla. Ortho- graphic feature transform for monocular 3d object detection. arXiv preprint arXiv:1811.08188, 2018. 7

    2018 arXiv

  28. [36]

    Cyclical learning rates for training neural networks

    Leslie N Smith. Cyclical learning rates for training neural networks. In 2017 IEEE winter conference on applications of computer vision (WACV), pages 464–472. IEEE, 2017. 5

    2017

  29. [37]

    Mo- tionnet: Joint perception and motion prediction for au- tonomous driving based on bird’s eye view maps

    Pengxiang Wu, Siheng Chen, and Dimitris N Metaxas. Mo- tionnet: Joint perception and motion prediction for au- tonomous driving based on bird’s eye view maps. In Pro- ceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 11385–11395, 2020. 1

    2020

  30. [38]

    M2bev: Multi-camera joint 3d detection and segmentation with unified birds-eye view representation

    Enze Xie, Zhiding Yu, Daquan Zhou, Jonah Philion, Anima Anandkumar, Sanja Fidler, Ping Luo, and Jose M Alvarez. M2bev: Multi-camera joint 3d detection and segmentation with unified birds-eye view representation. arXiv preprint arXiv:2204.05088, 2022. 1

    2022 arXiv

  31. [39]

    Auxadapt: Stable and efficient test-time adaptation for temporally consistent video semantic segmentation

    Yizhe Zhang, Shubhankar Borse, Hong Cai, and Fatih Porikli. Auxadapt: Stable and efficient test-time adaptation for temporally consistent video semantic segmentation. In Proceedings of the IEEE/CVF Winter Conference on Appli- cations of Computer Vision, pages 2339–2348, 2022. 2

    2022

  32. [40]

    Perceptual consistency in video segmentation

    Yizhe Zhang, Shubhankar Borse, Hong Cai, Ying Wang, Ning Bi, Xiaoyun Jiang, and Fatih Porikli. Perceptual consistency in video segmentation. In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, pages 2564–2573, 2022. 2

    2022

  33. [41]

    Beverse: Unified per- ception and prediction in birds-eye-view for vision-centric autonomous driving

    Yunpeng Zhang, Zheng Zhu, Wenzhao Zheng, Junjie Huang, Guan Huang, Jie Zhou, and Jiwen Lu. Beverse: Unified per- ception and prediction in birds-eye-view for vision-centric autonomous driving. arXiv preprint arXiv:2205.09743 ,

    arXiv

  34. [42]

    Dominant shuffle: A simple yet powerful data augmentation for time- series prediction

    Kai Zhao, Zuojie He, Alex Hung, and Dan Zeng. Dominant shuffle: A simple yet powerful data augmentation for time- series prediction. arXiv preprint arXiv:2405.16456 , 2024. 3

    2024 arXiv

  35. [43]

    Cross-view transform- ers for real-time map-view semantic segmentation

    Brady Zhou and Philipp Kr ¨ahenb¨uhl. Cross-view transform- ers for real-time map-view semantic segmentation. In Pro- ceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 13760–13769, 2022. 1, 7

    2022

  36. [44]

    Class-balanced grouping and sampling for point cloud 3d object detection

    Benjin Zhu, Zhengkai Jiang, Xiangxin Zhou, Zeming Li, and Gang Yu. Class-balanced grouping and sampling for point cloud 3d object detection. arXiv preprint arXiv:1908.09492,

    1908 arXiv

  37. [45]

    Monocular 3d vehicle de- tection using uncalibrated traffic cameras through homogra- phy

    Minghan Zhu, Songan Zhang, Yuanxin Zhong, Pingping Lu, Huei Peng, and John Lenneman. Monocular 3d vehicle de- tection using uncalibrated traffic cameras through homogra- phy. In 2021 IEEE/RSJ International Conference on Intel- ligent Robots and Systems (IROS), pages 3814–3821. IEEE,

    2021

Pith tools

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