Pith. sign in

REVIEW 3 major objections 6 minor 62 references

MambaFusion: Height-Fidelity Dense Global Fusion for Multi-modal 3D Object Detection

T0 review · 3 major / 6 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read MambaFusion claims the first pure-Mamba fusion block to reach state-of-the-art camera-LiDAR 3D object detection, with 75.0 NDS on nuScenes validation.

desk verdict Solid empirical paper with a real but under-isolated mechanism claim; worth refereeing with requests for proper ablations and code. read the letter →

arxiv 2507.04369 v1 pith:LX2SZKRR submitted 2025-07-06 cs.CV

Hanshi Wang , Jin Gao , Weiming Hu , Zhipeng Zhang This is my paper
classification cs.CV
keywords 3Dobjectdetectionmulti-modalfusionMambastatespacemodelcamera-LiDARheight-fidelityencodingdenseglobalnuScenes
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 sets out to show that a fusion module built purely from linear-complexity state-space models, specifically Mamba, can outperform windowed-transformer fusion for camera-LiDAR 3D object detection, at lower inference cost. The starting puzzle is that swapping Mamba into an existing fusion framework degrades accuracy, and the paper attributes the drop to height information being lost when LiDAR voxels are compressed into a shared coordinate space, which corrupts the order in which Mamba reads the fusion sequence. The proposed remedy is a height-fidelity LiDAR encoding that computes voxel coordinates in continuous space and filters out conflict-prone generated voxels, followed by a Hybrid Mamba Block that mixes local-window and bidirectional global Mamba over a Hilbert-curve serialization. The reported result is a nuScenes validation NDS of 75.0, surpassing previous best methods even those using higher-resolution images, and roughly 1.5x faster inference than a recent top method.

What carries the argument

The load-bearing mechanism is the pairing of height-fidelity LiDAR encoding with the Hybrid Mamba Block. Height-fidelity encoding replaces the discrete-space centroid of each voxel with a continuous-space ScatterMean over the raw point coordinates inside that voxel, and adds a conflict test that suppresses newly generated voxels whose merged position would be ambiguous; this keeps the z-coordinate precise enough that LiDAR features projected into image space land on the objects they describe. The Hybrid Mamba Block then processes these aligned features at two scales: a local Mamba over non-overlapping windows captures fine structure, and a bidirectional global Mamba over tokens serialized with a Hilbert curve, a space-filling curve that preserves spatial locality, captures full-scene context. The block is placed in a shared modality-aligner role, in raw-space fusion, and in BEV fusion, so it both harmonizes the two modalities and performs the final fused representation.

What would settle it

Compare the continuous-space height-fidelity encoding against discrete-space coordinates with identical model capacity and training schedule; if the discrete version with extra height channels or height-aware positional embedding reaches the same NDS, then coordinate precision is not the active ingredient. Conversely, randomize the Hilbert ordering while keeping height-fidelity coordinates; if performance does not collapse, sequence order is not what carries the gain.

Watch

Extended reading notes

Core claim

The central claim is that pure linear operations for fusion, not only quadratic attention, can reach state-of-the-art performance in camera-LiDAR 3D detection. The paper's key discovery is that the failure of vanilla Mamba in this setting is caused by height-information loss during multimodal alignment, and that this loss is fixable: by computing voxel coordinates through scatter-mean averaging in continuous 3D space rather than quantized discrete voxel grids, and by rejecting generated voxels that would merge ambiguously with neighbors, the LiDAR features project onto image features with far fewer mismatches. Once alignment is accurate, a Hybrid Mamba Block, composed of local Mamba for fine detail and bidirectional global Mamba for whole-scene context with tokens ordered along a Hilbert curve, delivers dense global fusion. On nuScenes this yields 75.0 NDS and 72.7 mAP on validation and 75.9 NDS on test, beating UniTR, IS-FUSION, and SparseLIF while running substantially faster.

Load-bearing premise

The paper's causal explanation, that height-information loss rather than optimization instability or the serialization itself is what makes vanilla Mamba degrade, is an attribution supported by ablations but not isolated in a controlled experiment that fixes capacity, optimizer, and ordering.

Editorial extensions

If this is right

  • Pure linear fusion blocks can replace windowed transformers inside a BEV detector without sacrificing accuracy, so latency-critical autonomy systems can spend the saved compute on other stages.
  • Height-fidelity encoding transfers: applying it to the windowed-transformer variant also improves performance, meaning the alignment fix is not specific to Mamba.
  • The fused representation generalizes beyond detection: plugging the method into FusionAD reduces average displacement error from 0.81 to 0.70 and collision rate from 0.12 to 0.10.
  • Robustness to degraded sensors improves, with the largest gain under unsynchronized LiDAR-camera timestamps, where global context compensates for broken local alignment.
  • Because the Hybrid Mamba Block is compatible with RetNet and RWKV as well, the design is a template for linear-attention fusion rather than a Mamba-only trick.

Reading between the lines

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

  • The causal story is not isolated: if optimization instability of SSM layers, rather than height loss, drives vanilla Mamba's drop, the height-fidelity encoding might still help without being the load-bearing mechanism; a controlled capacity-matched comparison would separate these.
  • Continuous-space coordinate fixes could plausibly improve other projection-based fusion tasks, such as radar-camera, LiDAR-maps, or point-text alignment, wherever discrete quantization corrupts correspondences.
  • The conflict test removes voxels likely to merge, which also reduces point density in cluttered regions; a density-matched baseline would reveal whether part of the gain is de-noising rather than height preservation.
  • A testable extension is to evaluate with higher-resolution images: if the margin over windowed transformers shrinks, height-fidelity matters most when image-LiDAR alignment is coarse; if it grows, the benefit lies in the global modeling itself.
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

3 major / 6 minor

Summary. The paper presents MambaFusion, a camera-LiDAR 3D object detection framework in which fusion is performed entirely with Mamba/linear-attention blocks rather than windowed transformers. The authors propose a Hybrid Mamba Block that combines local and global Mamba stages, and a Height-Fidelity LiDAR Encoding (HFL) that computes voxel coordinates in continuous space to preserve height information during multi-modal alignment. On nuScenes, the method reports 75.0 NDS / 72.7 mAP on validation and 75.9 NDS / 73.2 mAP on test with 4.7 FPS, surpassing prior methods including IS-FUSION and SparseLIF. The paper also reports ablations, robustness studies under sensor degradation, and an integration into the FusionAD planner.

Significance. If the reported results are reproducible, the paper makes a meaningful advance: it demonstrates that pure Mamba-based fusion can reach state-of-the-art accuracy on a standard multi-modal 3D detection benchmark while being faster than recent transformer-based competitors. The central conceptual contribution is the height-fidelity LiDAR encoding, which is claimed to fix a height-information loss that degrades vanilla Mamba. The paper is generally well written and the experimental suite is broad, including component ablations, robustness to sensor degradation, ERF visualizations, and an end-to-end planning evaluation. I credit the authors for releasing code and for comparing against external benchmarks. However, the load-bearing mechanistic claim about height information is not isolated in the experiments, and the ablation results are single-run, half-dataset numbers with small deltas. The benchmark result could still be correct, but the reason it works is underdetermined as presented.

major comments (3)
  1. [§3.4, Eqs. (8)–(10), Tab. 7] The paper's central claim is that height-information loss causes the degradation of vanilla Mamba and that Height-Fidelity LiDAR Encoding fixes it. However, the discrete-to-continuous change implemented by Eqs. (8)–(9) replaces voxel centroids with ScatterMean coordinates in all three axes, not only z, and Eq. (10) changes the set of generated voxels. These changes also alter the Hilbert indices in Eq. (3) and the local region partitions in Eq. (6), so the improvement from row ⑥ to row ⑦ in Tab. 7 (and from ③ to ⑤ in Tab. 4) is not attributable to height fidelity alone. I would need an ablation in which only the z coordinate is computed continuously while x and y remain on the discrete grid (or an equivalent isolation of height information) to support the height-specific mechanism.
  2. [§1, §4.3.4, Tab. 2] The causal story in Sec. 1—that Mamba's performance drop is caused by height-information loss during multi-modal alignment, 'leading to deviations in sequence order'—is asserted rather than demonstrated. The comparison in Tab. 7 from row ④ to row ⑦ changes the architecture (vanilla Mamba vs. hybrid local/global Mamba), the fusion space (Image+Frustum vs. Image+BEV), and the coordinate discretization simultaneously. No experiment holds model capacity, optimizer, sequence ordering, or architecture fixed while correcting only height information. Please add such an experiment, or explicitly weaken the causal claim to a design observation rather than a demonstrated mechanism.
  3. [§4.3, Tabs. 4–6] All component ablations are performed on half of the nuScenes validation set with a single run, and the reported deltas are small (e.g., 71.4→71.9 mAP and 73.6→74.3 NDS in Tab. 4 between ⑥ and ⑦; 71.5→71.9 mAP in Tab. 5(a)). Without multiple seeds or error bars, these differences may be within run-to-run variation, which undermines the component-level conclusions and the claim that HFL is the key enabler. Please report mean and standard deviation over at least three seeds for the main ablations, or clearly present the numbers as preliminary with the attendant uncertainty.
minor comments (6)
  1. [Abstract and Sec. 2] There are typos: 'top-tire NDS' should be 'top-tier NDS', and 'adapt pure linear attnetion' should be 'adapt pure linear attention'.
  2. [§3.4, Eq. (10)] The notation for the conflict test in Eq. (10) is hard to parse; please rewrite it with explicit set-builder notation and define the membership condition '∈ CP_s' more clearly.
  3. [§4.2, Tab. 3] The '1.5× faster' claim relative to IS-FUSION is ambiguous: Tab. 3 gives 5.4 FPS for MambaFusion-Lite (a 1.69× speedup) and 4.7 FPS for MambaFusion-Base (a 1.47× speedup) against IS-FUSION's 3.2 FPS; please specify which configuration is being compared.
  4. [§3.4, Fig. 5] Figure 5 is qualitative; reporting a quantitative projection or height error before and after HFL would directly support the height-loss narrative and complement the mAP/NDS tables.
  5. [§3.1, Tab. 2] The 'vanilla' linear-attention baseline in Tab. 2 is only described as replacing UniTR's fusion module; please clarify whether the exact same tokenizers, training schedule, and fusion space are used across all rows so that the comparison to UniTR is clean.
  6. [§4.3.1, Tab. 4] The table's checkmarks are not defined there; please add a legend defining Baseline-LC, HMB-M, HMB-R, HMB-B, and HFL in the caption for readability.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the central claims are benchmark-backed empirical results, and no load-bearing step reduces to its own inputs.

full rationale

The paper's main claim is that a pure Mamba-based fusion block with height-fidelity LiDAR encoding achieves an NDS of 75.0 on the nuScenes validation set. This is evaluated against an external benchmark and against existing methods such as UniTR, IS-FUSION, and SparseLIF, so it does not reduce to a fitted parameter or to the paper's own definitions. The height-fidelity encoding in Eqs. (8)-(10) computes voxel coordinates from point clouds in continuous space; it is a proposed mechanism, not a quantity defined in terms of the reported NDS or mAP. The Hybrid Mamba Block in Eqs. (3)-(7) is a concrete architecture, and the ablation tables (Tabs. 4-7) report measured differences rather than algebraic identities. There is no load-bearing self-citation: the cited works on Mamba, Hilbert curves, and LION are external, and the authors do not invoke their own prior results to justify the core design. The skeptic's concern that the continuous-coordinate change alters x, y, and z together, so the height attribution is underdetermined, is a correctness or attribution concern, not circularity: it does not show that any prediction is equivalent to an input by construction. Accordingly, the appropriate circularity finding is no significant circularity.

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

The central claim is empirical, so the ledger lists the architectural and dataset assumptions that the reported benchmark numbers depend on. No new physical entities, forces, or conserved quantities are introduced; the contributions are architectural techniques.

free parameters (4)
  • voxel resolution = [0.3, 0.3, 0.25] metres
    Chosen to match the UniTR baseline and affects the granularity of voxel compression and height-fidelity encoding.
  • local window size w = not reported in main text
    Determines the scale of local Mamba regions; selected by ablations but the exact value is omitted.
  • training epochs = 10
    Training duration; final benchmark numbers depend on this schedule and the one-cycle learning rate policy.
  • spatial range = [-54, 54] x [-54, 54] x [-5, 3] metres
    Defines the scene extent used for voxelization and fusion; standard for nuScenes experiments.
assumptions (4)
  • domain assumption nuScenes ground-truth annotations and sensor calibration are accurate enough for evaluating 3D detection.
    All metrics are computed against this benchmark dataset in Sec. 4.1.
  • domain assumption Hilbert curve serialization preserves spatial locality sufficiently for Mamba to model 3D interactions without explicit geometric attention.
    Global Mamba reorders tokens by Hilbert index in Eq. (3), a design premise that is not proven independently.
  • ad hoc to paper The performance drop of vanilla Mamba in Tab. 2 is caused by height-information loss during alignment.
    The paper asserts this attribution in Sec. 1 but does not isolate it experimentally.
  • domain assumption Linear-time state-space models can substitute for attention in fusion without losing necessary context.
    Core architectural premise of the Hybrid Mamba Block.

how reviews work

0 comments
Cite this review

Pith. "Pith review of MambaFusion: Height-Fidelity Dense Global Fusion for Multi-modal 3D Object Detection." pith.science (2026). https://pith.science/paper/LX2SZKRR

@misc{pith2026250704369,
  author       = {Pith},
  title        = {Pith review of: MambaFusion: Height-Fidelity Dense Global Fusion for Multi-modal 3D Object Detection},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/LX2SZKRR}},
  note         = {Machine review of arXiv:2507.04369}
}
read the original abstract

We present the first work demonstrating that a pure Mamba block can achieve efficient Dense Global Fusion, meanwhile guaranteeing top performance for camera-LiDAR multi-modal 3D object detection. Our motivation stems from the observation that existing fusion strategies are constrained by their inability to simultaneously achieve efficiency, long-range modeling, and retaining complete scene information. Inspired by recent advances in state-space models (SSMs) and linear attention, we leverage their linear complexity and long-range modeling capabilities to address these challenges. However, this is non-trivial since our experiments reveal that simply adopting efficient linear-complexity methods does not necessarily yield improvements and may even degrade performance. We attribute this degradation to the loss of height information during multi-modal alignment, leading to deviations in sequence order. To resolve this, we propose height-fidelity LiDAR encoding that preserves precise height information through voxel compression in continuous space, thereby enhancing camera-LiDAR alignment. Subsequently, we introduce the Hybrid Mamba Block, which leverages the enriched height-informed features to conduct local and global contextual learning. By integrating these components, our method achieves state-of-the-art performance with the top-tire NDS score of 75.0 on the nuScenes validation benchmark, even surpassing methods that utilize high-resolution inputs. Meanwhile, our method maintains efficiency, achieving faster inference speed than most recent state-of-the-art methods.

Figures

Figures reproduced from arXiv: 2507.04369 by the authors.

Figure 1
Figure 1. Illustration of the LiDAR and camera fusion framework: [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Overview of our framework. We propose a novel height-fidelity LiDAR encoding method, which preserves more accurate height information within voxel features, thereby enhancing the precision of multi-modal alignment. Then, we introduce the Hybrid Mamba Block (HMB), designed to: 1) align the latent-space distributions across modalities, 2) facilitate cross-modal information exchange between raw voxel and camera feature… view at source ↗
Figure 3
Figure 3. Illustration of the Hybrid Mamba Block. The block first employs the Local Mamba to enable localized feature in￾teraction around each spatial position. Subsequently, the Global Mamba facilitates interaction across the entire space. 3.3. Hybrid Mamba Block To achieve dense global fusion, we introduce the Hybrid Mamba Block, which combines local and global Mamba modules to integrate information across all modalities in… view at source ↗
Figures from the paper (3 more)
Figure 5
Figure 5. Figure 5: Voxel projection and detection results on images. Comparison of our projection method (bottom), enhanced by our Height￾Fidelity LiDAR encoding, with previous methods (top). The figure demonstrates that projection errors introduce false positives, category errors, and u…
Figure 6
Figure 6. Figure 6: The effective receptive fields (ERFs). “Win” de￾notes window partition, “Hybrid” means Hybrid Mamba (RetNet, RWKV) Block, “HFL” refers to Height-Fidelity LiDAR Encoding. incorporate the Hybrid Mamba Block along with the height￾fidelity LiDAR encoding (⑥ and ⑦). These m…
Figure 7
Figure 7. Figure 7: Qualitative comparison of different fusion strategies. The green, blue and red boxes indicate ground truth, output of previous methods, and our results, respectively. locations. We observe that UniTR [47] with a window￾based transformer captures local information, limi…

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

62 extracted references · 41 canonical work pages

  1. [1]

    Transfusion: Robust lidar-camera fusion for 3d object detection with transform- ers

    Xuyang Bai, Zeyu Hu, Xinge Zhu, Qingqiu Huang, Yilun Chen, Hongbo Fu, and Chiew-Lan Tai. Transfusion: Robust lidar-camera fusion for 3d object detection with transform- ers. In CVPR, pages 1090–1099, 2022. 6, 8

    work page 2022

  2. [2]

    nuscenes: A mul- timodal dataset for autonomous driving

    Holger Caesar, Varun Bankiti, Alex H Lang, Sourabh V ora, Venice Erin Liong, Qiang Xu, Anush Krishnan, Yu Pan, Giancarlo Baldan, and Oscar Beijbom. nuscenes: A mul- timodal dataset for autonomous driving. In CVPR, pages 11621–11631, 2020. 1, 2, 5, 6

    work page 2020

  3. [3]

    Objectfusion: Multi-modal 3d object detection with object-centric fusion

    Qi Cai, Yingwei Pan, Ting Yao, Chong-Wah Ngo, and Tao Mei. Objectfusion: Multi-modal 3d object detection with object-centric fusion. In ICCV, pages 18067–18076, 2023. 6

    work page 2023

  4. [4]

    Deformable feature aggrega- tion for dynamic multi-modal 3d object detection

    Zehui Chen, Zhenyu Li, Shiquan Zhang, Liangji Fang, Qin- hong Jiang, and Feng Zhao. Deformable feature aggrega- tion for dynamic multi-modal 3d object detection. In ECCV, pages 628–644. Springer, 2022. 6

    work page 2022

  5. [5]

    Transformers are ssms: General- ized models and efficient algorithms through structured state space duality

    Tri Dao and Albert Gu. Transformers are ssms: General- ized models and efficient algorithms through structured state space duality. arXiv preprint arXiv:2405.21060, 2024. 4

    arXiv 2024

  6. [6]

    Scaling up your kernels to 31x31: Revisiting large kernel design in cnns

    Xiaohan Ding, Xiangyu Zhang, Jungong Han, and Guiguang Ding. Scaling up your kernels to 31x31: Revisiting large kernel design in cnns. In CVPR, pages 11963–11975, 2022. 2

    work page 2022

  7. [7]

    Mambast: A plug-and-play cross-spectral spatial- temporal fuser for efficient pedestrian detection

    Xiangbo Gao, Asiegbu Miracle Kanu-Asiegbu, and Xiaox- iao Du. Mambast: A plug-and-play cross-spectral spatial- temporal fuser for efficient pedestrian detection. arXiv preprint arXiv:2408.01037, 2024. 3

    arXiv 2024

  8. [8]

    Mamba: Linear-time sequence modeling with selective state spaces

    Albert Gu and Tri Dao. Mamba: Linear-time sequence modeling with selective state spaces. arXiv preprint arXiv:2312.00752, 2023. 1, 3, 4

    arXiv 2023

Show all 62 references
  1. [9]

    Hippo: Recurrent memory with optimal polynomial projections

    Albert Gu, Tri Dao, Stefano Ermon, Atri Rudra, and Christo- pher R´e. Hippo: Recurrent memory with optimal polynomial projections. NIPS, 33:1474–1487, 2020. 3, 4

    2020

  2. [10]

    Efficiently modeling long sequences with structured state spaces

    Albert Gu, Karan Goel, and Christopher R ´e. Efficiently modeling long sequences with structured state spaces. arXiv preprint arXiv:2111.00396, 2021. 4

    2021 arXiv

  3. [11]

    Combining recurrent, convolutional, and continuous-time models with linear state space layers

    Albert Gu, Isys Johnson, Karan Goel, Khaled Saab, Tri Dao, Atri Rudra, and Christopher R ´e. Combining recurrent, convolutional, and continuous-time models with linear state space layers. NeurIPS, 34:572–585, 2021. 4

    2021

  4. [12]

    Exploring recurrent long-term temporal fusion for multi-view 3d perception

    Chunrui Han, Jinrong Yang, Jianjian Sun, Zheng Ge, Run- pei Dong, Hongyu Zhou, Weixin Mao, Yuang Peng, and Xi- angyu Zhang. Exploring recurrent long-term temporal fusion for multi-view 3d perception. 2024. 2

    2024

  5. [13]

    Demystify mamba in vision: A linear attention perspective

    Dongchen Han, Ziyi Wang, Zhuofan Xia, Yizeng Han, Yi- fan Pu, Chunjiang Ge, Jun Song, Shiji Song, Bo Zheng, and Gao Huang. Demystify mamba in vision: A linear attention perspective. NIPS, 37:127181–127203, 2025. 3

    2025

  6. [14]

    Linear systems theory

    Joao P Hespanha. Linear systems theory. Princeton univer- sity press, 2018. 4

    2018

  7. [15]

    ¨Uber die stetige abbildung einer linie auf ein fl ¨achenst¨uck

    David Hilbert and David Hilbert. ¨Uber die stetige abbildung einer linie auf ein fl ¨achenst¨uck. Dritter Band: Analysis· Grundlagen der Mathematik· Physik Verschiedenes: Nebst Einer Lebensgeschichte, pages 1–2, 1935. 4, 7

    1935

  8. [16]

    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

  9. [17]

    Detecting as labeling: Rethinking lidar-camera fusion in 3d object detection

    Junjie Huang, Yun Ye, Zhujin Liang, Yi Shan, and Dalong Du. Detecting as labeling: Rethinking lidar-camera fusion in 3d object detection. In ECCV, pages 439–455. Springer,

  10. [18]

    Ep- net: Enhancing point features with image semantics for 3d object detection

    Tengteng Huang, Zhe Liu, Xiwu Chen, and Xiang Bai. Ep- net: Enhancing point features with image semantics for 3d object detection. In ECCV, pages 35–52. Springer, 2020. 2

    2020

  11. [19]

    Ccnet: Criss-cross attention for semantic segmentation

    Zilong Huang, Xinggang Wang, Lichao Huang, Chang Huang, Yunchao Wei, and Wenyu Liu. Ccnet: Criss-cross attention for semantic segmentation. In ICCV, pages 603– 612, 2019. 3

    2019

  12. [20]

    Transformers are rnns: Fast autoregressive transformers with linear attention

    Angelos Katharopoulos, Apoorv Vyas, Nikolaos Pappas, and Franc ¸ois Fleuret. Transformers are rnns: Fast autoregressive transformers with linear attention. In IMCL, pages 5156–

  13. [21]

    Unifying voxel-based representation with transformer for 3d object detection

    Yanwei Li, Yilun Chen, Xiaojuan Qi, Zeming Li, Jian Sun, and Jiaya Jia. Unifying voxel-based representation with transformer for 3d object detection. NeurIPS, 35:18442– 18455, 2022. 6

    2022

  14. [22]

    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 AAAI, pages 1477–1485, 2023. 2

    2023

  15. [23]

    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, Yu Qiao, and Jifeng Dai. Bevformer: Learning bird’s-eye-view representation from multi-camera images via spatiotemporal transformers. In ECCV, pages 1–

  16. [24]

    Mamba24/8d: Enhancing global interaction in point clouds via state space model

    Zhuoyuan Li, Yubo Ai, Jiahao Lu, ChuXin Wang, Jiacheng Deng, Hanzhi Chang, Yanzhe Liang, Wenfei Yang, Shifeng Zhang, and Tianzhu Zhang. Mamba24/8d: Enhancing global interaction in point clouds via state space model. arXiv preprint arXiv:2406.17442, 2024. 4

    2024 arXiv

  17. [25]

    Point- mamba: A simple state space model for point cloud analysis

    Dingkang Liang, Xin Zhou, Xinyu Wang, Xingkui Zhu, Wei Xu, Zhikang Zou, Xiaoqing Ye, and Xiang Bai. Point- mamba: A simple state space model for point cloud analysis. arXiv preprint arXiv:2402.10739, 2024. 3

    2024 arXiv

  18. [26]

    Bevfusion: A simple and robust lidar-camera fusion framework

    Tingting Liang, Hongwei Xie, Kaicheng Yu, Zhongyu Xia, Zhiwei Lin, Yongtao Wang, Tao Tang, Bing Wang, and Zhi Tang. Bevfusion: A simple and robust lidar-camera fusion framework. NeurIPS, 35:10421–10434, 2022. 1, 2, 6, 8

    2022

  19. [27]

    Sparse4d v2: Recurrent temporal fusion with sparse model

    Xuewu Lin, Tianwei Lin, Zixiang Pei, Lichao Huang, and Zhizhong Su. Sparse4d v2: Recurrent temporal fusion with sparse model. arXiv preprint arXiv:2305.14018, 2023. 2

    2023 arXiv

  20. [28]

    Sparsebev: High-performance sparse 3d object de- tection from multi-camera videos

    Haisong Liu, Yao Teng, Tao Lu, Haiguang Wang, and Limin Wang. Sparsebev: High-performance sparse 3d object de- tection from multi-camera videos. In ICCV, pages 18580– 18590, 2023. 2

    2023

  21. [29]

    More convnets in the 2020s: Scaling up kernels beyond 51x51 using sparsity

    Shiwei Liu, Tianlong Chen, Xiaohan Chen, Xuxi Chen, Qiao Xiao, Boqian Wu, Tommi K¨arkk¨ainen, Mykola Pechenizkiy, Decebal Mocanu, and Zhangyang Wang. More convnets in the 2020s: Scaling up kernels beyond 51x51 using sparsity. arXiv preprint arXiv:2207.03620, 2022. 2

    2022 arXiv

  22. [30]

    Petr: Position embedding transformation for multi-view 3d object detection

    Yingfei Liu, Tiancai Wang, Xiangyu Zhang, and Jian Sun. Petr: Position embedding transformation for multi-view 3d object detection. In ECCV, pages 531–548. Springer, 2022. 2

    2022

  23. [31]

    Vmamba: Visual state space model

    Yue Liu, Yunjie Tian, Yuzhong Zhao, Hongtian Yu, Lingxi Xie, Yaowei Wang, Qixiang Ye, and Yunfan Liu. Vmamba: Visual state space model. arXiv preprint arXiv:2401.10166,

    arXiv

  24. [32]

    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. pages 2774–2781. IEEE, 2023. 1, 2, 6, 8

    2023

  25. [33]

    Lion: Linear group rnn for 3d object detection in point clouds

    Zhe Liu, Jinghua Hou, Xinyu Wang, Xiaoqing Ye, Jingdong Wang, Hengshuang Zhao, and Xiang Bai. Lion: Linear group rnn for 3d object detection in point clouds. arXiv preprint arXiv:2407.18232, 2024. 3, 4, 5, 6

    2024 arXiv

  26. [34]

    A computer oriented geodetic data base and a new technique in file sequencing

    Guy M Morton. A computer oriented geodetic data base and a new technique in file sequencing. 1966. 7

    1966

  27. [35]

    Rwkv: Reinventing rnns for the transformer era

    Bo Peng, Eric Alcaide, Quentin Anthony, Alon Albalak, Samuel Arcadinho, Stella Biderman, Huanqi Cao, Xin Cheng, Michael Chung, Matteo Grella, et al. Rwkv: Reinventing rnns for the transformer era. arXiv preprint arXiv:2305.13048, 2023. 1, 3, 8

    2023 arXiv

  28. [36]

    Lift, splat, shoot: Encoding images from arbitrary camera rigs by implicitly unprojecting to 3d

    Jonah Philion and Sanja Fidler. Lift, splat, shoot: Encoding images from arbitrary camera rigs by implicitly unprojecting to 3d. In ECCV, pages 194–210. Springer, 2020. 2

    2020

  29. [37]

    Pointnet++: Deep hierarchical feature learning on point sets in a metric space

    Charles Ruizhongtai Qi, Li Yi, Hao Su, and Leonidas J Guibas. Pointnet++: Deep hierarchical feature learning on point sets in a metric space. NeurIPS, 30, 2017. 2

    2017

  30. [38]

    Hornet: Efficient high- order spatial interactions with recursive gated convolutions

    Yongming Rao, Wenliang Zhao, Yansong Tang, Jie Zhou, Ser Nam Lim, and Jiwen Lu. Hornet: Efficient high- order spatial interactions with recursive gated convolutions. NeurIPS, 35:10353–10366, 2022. 2

    2022

  31. [39]

    Mvx- net: Multimodal voxelnet for 3d object detection

    Vishwanath A Sindagi, Yin Zhou, and Oncel Tuzel. Mvx- net: Multimodal voxelnet for 3d object detection. pages 7276–7282. IEEE, 2019. 2, 8

    2019

  32. [40]

    Programming the hilbert curve

    John Skilling. Programming the hilbert curve. In AIP Con- ference Proceedings, pages 381–387. American Institute of Physics, 2004. 4

    2004

  33. [41]

    Cyclical learning rates for training neural networks

    Leslie N Smith. Cyclical learning rates for training neural networks. pages 464–472. IEEE, 2017. 6

    2017

  34. [42]

    Scalability in perception for autonomous driving: Waymo open dataset

    Pei Sun, Henrik Kretzschmar, Xerxes Dotiwalla, Aurelien Chouard, Vijaysai Patnaik, Paul Tsui, James Guo, Yin Zhou, Yuning Chai, Benjamin Caine, et al. Scalability in perception for autonomous driving: Waymo open dataset. In CVPR, pages 2446–2454, 2020. 8

    2020

  35. [43]

    Retentive network: A successor to transformer for large language mod- els

    Yutao Sun, Li Dong, Shaohan Huang, Shuming Ma, Yuqing Xia, Jilong Xue, Jianyong Wang, and Furu Wei. Retentive network: A successor to transformer for large language mod- els. arXiv preprint arXiv:2307.08621, 2023. 1, 3, 8

    2023 arXiv

  36. [44]

    Attention is all you need

    A Vaswani. Attention is all you need. NeurIPS, 2017. 2

    2017

  37. [45]

    Pointpainting: Sequential fusion for 3d object detec- tion

    Sourabh V ora, Alex H Lang, Bassam Helou, and Oscar Bei- jbom. Pointpainting: Sequential fusion for 3d object detec- tion. In CVPR, pages 4604–4612, 2020. 2, 6

    2020

  38. [46]

    Pointaugmenting: Cross-modal augmentation for 3d object detection

    Chunwei Wang, Chao Ma, Ming Zhu, and Xiaokang Yang. Pointaugmenting: Cross-modal augmentation for 3d object detection. In CVPR, pages 11794–11803, 2021. 2, 6, 8

    2021

  39. [47]

    Unitr: A unified and efficient multi-modal transformer for bird’s-eye-view repre- sentation

    Haiyang Wang, Hao Tang, Shaoshuai Shi, Aoxue Li, Zhen- guo Li, Bernt Schiele, and Liwei Wang. Unitr: A unified and efficient multi-modal transformer for bird’s-eye-view repre- sentation. In ICCV, pages 6792–6802, 2023. 2, 3, 6, 7, 8

    2023

  40. [48]

    Fcos3d: Fully convolutional one-stage monocular 3d object detection

    Tai Wang, Xinge Zhu, Jiangmiao Pang, and Dahua Lin. Fcos3d: Fully convolutional one-stage monocular 3d object detection. In ICCV, pages 913–922, 2021. 2

    2021

  41. [49]

    Detr3d: 3d object detection from multi-view images via 3d-to-2d queries

    Yue Wang, Vitor Campagnolo Guizilini, Tianyuan Zhang, Yilun Wang, Hang Zhao, and Justin Solomon. Detr3d: 3d object detection from multi-view images via 3d-to-2d queries. pages 180–191. PMLR, 2022. 2

    2022

  42. [50]

    Cross modal transformer via coordinates encoding for 3d object dectection

    Junjie Yan, Yingfei Liu, Jianjian Sun, Fan Jia, Shuailin Li, Tiancai Wang, and Xiangyu Zhang. Cross modal transformer via coordinates encoding for 3d object dectection. arXiv preprint arXiv:2301.01283, 2(3):4, 2023. 6

    2023 arXiv

  43. [51]

    Deepinteraction: 3d object detection via modality interaction

    Zeyu Yang, Jiaqi Chen, Zhenwei Miao, Wei Li, Xiatian Zhu, and Li Zhang. Deepinteraction: 3d object detection via modality interaction. NeurIPS, 35:1992–2005, 2022. 6

    1992

  44. [52]

    Fusionad: Multi-modality fusion for prediction and planning tasks of autonomous driving

    Tengju Ye, Wei Jing, Chunyong Hu, Shikun Huang, Ling- ping Gao, Fangzhen Li, Jingke Wang, Ke Guo, Wencong Xiao, Weibo Mao, et al. Fusionad: Multi-modality fusion for prediction and planning tasks of autonomous driving. arXiv preprint arXiv:2308.01006, 2023. 8

    2023 arXiv

  45. [53]

    Is-fusion: Instance-scene collaborative fusion for multimodal 3d object detection

    Junbo Yin, Jianbing Shen, Runnan Chen, Wei Li, Ruigang Yang, Pascal Frossard, and Wenguan Wang. Is-fusion: Instance-scene collaborative fusion for multimodal 3d object detection. In CVPR, pages 14905–14915, 2024. 2, 3, 6

    2024

  46. [54]

    Multi- modal virtual point 3d detection.NeurIPS, 34:16494–16507,

    Tianwei Yin, Xingyi Zhou, and Philipp Kr ¨ahenb¨uhl. Multi- modal virtual point 3d detection.NeurIPS, 34:16494–16507,

  47. [55]

    Drama: An efficient end-to-end motion planner for autonomous driving with mamba

    Chengran Yuan, Zhanqi Zhang, Jiawei Sun, Shuo Sun, Ze- fan Huang, Christina Dao Wen Lee, Dongen Li, Yuhang Han, Anthony Wong, Keng Peng Tee, et al. Drama: An efficient end-to-end motion planner for autonomous driving with mamba. arXiv preprint arXiv:2408.03601, 2024. 3

    2024 arXiv

  48. [56]

    V oxel mamba: Group-free state space models for point cloud based 3d object detection

    Guowen Zhang, Lue Fan, Chenhang He, Zhen Lei, Zhaox- iang Zhang, and Lei Zhang. V oxel mamba: Group-free state space models for point cloud based 3d object detection. arXiv preprint arXiv:2406.10700, 2024. 3

    2024 arXiv

  49. [57]

    Sparselif: High-performance sparse lidar- camera fusion for 3d object detection

    Hongcheng Zhang, Liu Liang, Pengxin Zeng, Xiao Song, and Zhe Wang. Sparselif: High-performance sparse lidar- camera fusion for 3d object detection. In ECCV, pages 109–

  50. [58]

    Point could mamba: Point cloud learning via state space model

    Tao Zhang, Xiangtai Li, Haobo Yuan, Shunping Ji, and Shuicheng Yan. Point could mamba: Point cloud learning via state space model. arXiv preprint arXiv:2403.00762, 2024. 3

    2024 arXiv

  51. [59]

    Sparsefusion: Dis- tilling view-conditioned diffusion for 3d reconstruction

    Zhizhuo Zhou and Shubham Tulsiani. Sparsefusion: Dis- tilling view-conditioned diffusion for 3d reconstruction. In CVPR, pages 12588–12597, 2023. 2, 6, 8

    2023

  52. [60]

    Vision mamba: Efficient visual representation learning with bidirectional state space model

    Lianghui Zhu, Bencheng Liao, Qian Zhang, Xinlong Wang, Wenyu Liu, and Xinggang Wang. Vision mamba: Efficient visual representation learning with bidirectional state space model. arXiv preprint arXiv:2401.09417, 2024. 3

    2024 arXiv

  53. [61]

    Flatfusion: Delving into details of sparse transformer-based camera-lidar fusion for autonomous driving

    Yutao Zhu, Xiaosong Jia, Xinyu Yang, and Junchi Yan. Flatfusion: Delving into details of sparse transformer-based camera-lidar fusion for autonomous driving. arXiv preprint arXiv:2408.06832, 2024. 2

    2024 arXiv

  54. [128]

    Springer, 2025. 2, 6

    2025

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.