Mapping like a Skeptic: Probabilistic BEV Projection for Online HD Mapping

Fatih Erdo\u{g}an; Fatma G\"uney; Merve Rabia Bar{\i}n

REVIEW 3 major objections 5 minor 32 references

Mapping like a Skeptic: Probabilistic BEV Projection for Online HD Mapping

T0 review · 3 major / 5 minor · reviewed 2026-08-05 · deepseek-v4-flash

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

Pith's one-line read Replacing attention-based projection with a geometry-initialized, confidence-gated probabilistic sampler yields more accurate online HD maps with fewer hallucinated road elements, especially at long range.

desk verdict A useful BEV projection idea with a real attribution gap: the paper never isolates the probabilistic sampling from the learned offsets, so the central claim overreaches even though the direction is sound. read the letter →

arxiv 2508.21689 v1 pith:KNZXYBWN submitted 2025-08-29 cs.CV

Fatih Erdoğan , Merve Rabia Bar{i}n , Fatma Güney This is my paper

classification cs.CV

keywords onlineHDmappingbird's-eyeviewprojectionprobabilisticsamplingcamera-to-BEVtransformtemporalfusionautonomousdrivinggeneralizationconfidencegating

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

Online HD mapping for self-driving cars needs to move image evidence onto a bird's-eye-view grid, and this paper's claim is that the usual way of doing that — learned, attention-based projection — is the main source of error: it memorizes where road elements tend to appear and invents them at the wrong place. The authors propose to start from the explicit geometric mapping given by camera parameters and refine it scene-by-scene with a probabilistic sampler: each grid cell learns a small Gaussian over where in the image to pull features, and a confidence score per pixel filters out irrelevant content like pedestrians and walls. The same confidences gate temporal fusion, so only reliable historical features accumulate. On geographically disjoint splits of nuScenes and Argoverse2 the method beats the current best baselines on nearly every metric, with the biggest gains in the long perception range, which is why a reader should care: accuracy and generalization to unseen areas are precisely what online mapping has been missing.

What carries the argument

The central object is the probabilistic projection module. It starts from a static pull mapping defined by camera intrinsics and extrinsics, then predicts per-pixel offset distributions and covariances with small CNN heads — foffset receives the camera features plus a distance mask, and fconf predicts covariance and confidence. For each BEV grid cell i, the refined parameters define a Gaussian N(µ_i, Σ_i) over image sampling locations; K locations are sampled, and the cell's feature is built as a weighted sum of bilinearly sampled features, with each weight equal to the sampled pixel's confidence times the Gaussian likelihood of the sample. A second use of the same confidence scores implemen

What would settle it

Run the model on a sequence where a lane line is split by occlusion into two separated image regions: a single Gaussian per cell can sample only near one mode, so if the predicted map shows a gap or collapsed confidence where an attention baseline recovers both segments, the unimodal assumption is the binding limit. Separately, evaluating on the original overlapping nuScenes split would reveal how much of the reported margin depends on the region-disjoint evaluation the paper adopts.

Watch

Extended reading notes

Core claim

The paper establishes that the camera-to-BEV projection — not the vector-map head — is where online HD mapping loses accuracy. Replacing the deformable-attention projection inside the MapTracker pipeline with a plain geometric pull mapping from camera intrinsics and extrinsics already matches or slightly beats the original, which the authors read as evidence that attention's learned geometry memorizes the training distribution and hallucinates road elements where they usually appear. Their method then makes the projection probabilistic: a static BEV-to-image correspondence is refined by learned offsets, each BEV cell draws K samples from the resulting Gaussian N(µ_i, Σ_i), and the cell's fea

Load-bearing premise

Each BEV cell's true image source is assumed to be single-peaked — one Gaussian per cell — so K samples from that Gaussian can pull the right features; where the true source is split across separate image regions (occlusion, repeated textures) the model cannot represent it, which is why occluded targets remain the paper's admitted limitation.

Editorial extensions

If this is right

If the claim holds, attention-based BEV projection is not required for accurate online HD mapping; a geometry-seeded, scene-adapted sampler is more accurate and generalizes to unseen regions.
Long-range mapping (100 m × 50 m) benefits most, so the approach could extend effective perception range without adding sensors.
Confidence-gated temporal accumulation improves consistency-aware performance on nuScenes, implying the gating signal is a reusable mechanism for streaming perception.
Fewer false positives near junctions and crossings means downstream planning receives a cleaner map, not just a higher mAP.
The gains are dataset-dependent (large on nuScenes, modest on Argoverse2, slightly negative for the boundary class at short range on Argoverse2), so the method's advantage is strongest where attention baselines struggle.

Reading between the lines

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

A natural extension the authors leave open: replacing the single Gaussian per cell with a mixture or a sampled heatmap could recover occluded features that are split across disjoint image regions.
The confidence map is an interpretable by-product; a planner could consume it as per-cell evidence quality and treat low-confidence areas as untrusted before acting on them.
The static-pull baseline's surprisingly strong showing suggests other BEV lifting tasks (object detection, segmentation) may also be over-relying on learned projection when camera geometry is available.
A hybrid that keeps geometric sampling for most cells but activates context-aware attention only where confidence is low could address occlusion while preserving the geometry anchor.

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.

Desk Editor's Note

read the letter

Colleague,

Read this paper on probabilistic BEV projection for HD mapping. The idea is simple: instead of attention-based projection (BEVFormer-style) used by MapTracker, they start from known camera geometry, learn per-pixel offsets, and add a Gaussian sampling step with confidence weights. They also use confidences to merge historical BEV features temporally. That's a real, useful direction, and the paper shows consistent gains on nuScenes and long-range Argoverse2, with fewer false positives. The ablation in Table 2 is genuinely informative: replacing attention with static geometric pull already beats MapTracker slightly (41.0 vs 40.3), and learning offsets adds +2.5 mAP. Code and checkpoints are promised.

Now the soft spots. The biggest one is attribution. The "probabilistic" part—covariance, K samples, stochastic sampling—is never isolated. Table 2 only compares static mapping (A) to the full offset+covariance model (B1). There's no fixed-covariance or K=1 ablation, and no test of deterministic mean vs sampling at inference. So we don't know if the gains come from the Gaussian machinery or just from learned deterministic residual offsets. The stress-test concern lands: the paper's strongest claim, that probabilistic projection is what drives the improvements, is not yet supported by the evidence as reported. That's not fatal—the offsets alone are still a contribution—but the framing overreaches.

Second, the experimental comparisons are thin. Only two baselines, one taken as reported, and no error bars. On Argoverse2 the method slips on boundary AP and C-mAP in the short range, which the paper acknowledges but doesn't stress. The improvements on nuScenes are large, but a single run leaves some uncertainty about stability.

Third, the single-Gaussian assumption for correspondence uncertainty is a reasonable first cut but likely weak in occluded or repetitive-texture areas; the authors themselves note occluded regions are handled poorly.

Who is this for? People working on online HD mapping and BEV projection will get value. It deserves a serious referee: the direction is promising, the experiments are mostly clean, and the open questions are addressable with a few additional ablations and maybe repeated runs. I'd want the authors to do the K=1 / fixed-covariance ablation and clarify whether probabilistic sampling is actually load-bearing. If they can show that, this is a solid paper. If they can't, the claim needs to be scaled back to "learned residual offsets plus confidence-gated temporal fusion." Either way, worth engaging.

Recommendation: send to peer review, require the missing ablation before acceptance.

Referee Report

3 major / 5 minor

Summary. The paper proposes replacing the attention-based BEV projection used in online HD map estimation with a geometry-initialized, scene-adaptive probabilistic projection. Starting from a static camera-parameter pull mapping μ_pull, a lightweight CNN predicts per-pixel offsets and per-pixel covariance/confidence; for each BEV cell a Gaussian N(μ_i, Σ_i) is formed, K locations are sampled from it, and image features at those locations are pooled with likelihood/confidence weights (Eq. 6). Per-camera raw BEV features are averaged, then confidence-weighted temporal fusion with historical raw features and a MapTracker-style memory module produces the final BEV features. Experiments on new non-overlapping geographical splits of nuScenes and Argoverse2 report consistent mAP/C-mAP improvements over StreamMapNet and MapTracker, with the largest gains on nuScenes and at long range. Code and checkpoints are released.

Significance. If confirmed, the paper would make a useful empirical contribution: it shows that a geometric prior from camera extrinsics, combined with residual offset refinement and confidence-gated temporal accumulation, can improve generalization and reduce false positives relative to learned attention-based projection. The use of new geographical splits is a genuine generalization test, and the ablation structure is logical: A→B1 isolates offset/covariance refinement, C1→C2 isolates confidence-gated temporal fusion, and D→E isolates confidence weighting in history fusion. The release of code and checkpoints is a practical strength. However, the central attribution of the gains to probabilistic sampling is not yet supported: the ablations do not separate deterministic offset refinement from Gaussian sampling, and the uncertainty model itself is not evaluated. The reported comparisons also rest on single-run numbers with no variance estimates, and on Argoverse2 some metrics are actually lower than MapTracker. These issues make the central claim stronger than the current evidence.

major comments (3)

[§3.1 Eq. (6); §4.3 Table 2] The reported gain from A to B1 (41.0→43.5 mAP) conflates three mechanisms: the learned residual offset in Eqs. (3)–(4), the predicted covariance in Eq. (5), and stochastic K-sample pooling in Eq. (6). No row in Table 2 removes the Gaussian covariance while keeping the offset, no row fixes K=1 with deterministic sampling at the mean, and no row compares K=1 vs K>1. If the learned Σ collapses or the K samples concentrate at μ_i, Eq. (6) degenerates to deterministic bilinear sampling at a refined location, and the abstract's claim that 'probabilistic projection' drives the improvement is unsupported. Please add ablations for (i) deterministic sampling at μ_i without Σ, (ii) learned Σ but K=1/deterministic mean, and (iii) fixed isotropic Σ with K>1, and report learned covariance statistics to demonstrate that the stochastic component is actually active.
[§4, Table 1] All results appear to be from a single run, and no error bars or significance tests are reported. On Argoverse2 the short-range boundary AP is below MapTracker (67.5 vs 68.9) and short-range C-mAP is below MapTracker (62.7 vs 63.2); the long-range C-mAP gain is only 0.7 points. Without variance estimates, 'significantly outperforms' across 'nearly all metrics' is too strong. Please provide mean/std over multiple seeds or a paired significance test, and clarify exactly which Table 1 entries are re-runs versus numbers taken from prior papers.
[§3.1, Fig. 2, Limitations] The term 'probabilistic' is not validated as uncertainty modeling. The covariance Σ and confidence α are trained only through the downstream map loss; no calibration, entropy, or failure-mode analysis of the Gaussian is presented, and the Limitations paragraph concedes that occluded regions are handled poorly. If the contribution is intended to be probabilistic projection rather than merely offset-refined sampling, the revision should evaluate the uncertainty estimates (e.g., covariance behavior in occluded vs visible cells, or calibration of confidence) or soften the probabilistic framing.

minor comments (5)

[§3.2, Eq. (8)] The equation uses Braw on both sides without distinguishing the current and updated variables; make the warping and update explicit with time subscripts or a separate update operator.
[§3.1, Eq. (6)] The text says the Gaussian likelihood is normalized by the sum of K likelihoods, but the displayed equation omits this normalization; align the formula with the text.
[§4.3, Table 2 discussion] The sentence 'Removing confidence scores has the least impact on mAP (D)' is confusing because D vs E drops 2.0 mAP, which is comparable to the other ablated components; clarify the intended comparison.
[§4, Experimental Setup] C-mAP is used as a headline metric but is never defined; provide the formula or a precise citation.
[Fig. 2 caption] The note about image-space vs BEV-space notation for μ_pull and μ is difficult to parse; consider simplifying the notation or moving the clarification to the main text.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the reported gains are empirical comparisons on held-out splits, not consequences of fitted inputs or self-citations.

full rationale

The derivation chain is self-contained and empirically grounded. The core contribution (Eq. 6) is a defined probabilistic sampling mechanism: offsets and covariances are predicted from camera features and the BEV cell feature is a confidence-weighted sum over K samples. Nothing in this construction is defined in terms of the final AP numbers, and the evaluation compares against external baselines (StreamMapNet, MapTracker) on non-overlapping geographical splits. The improvements are observed results, not outputs of a formula that already contains them. There are no load-bearing self-citations: the cited projection (Simple-BEV) and attention baseline (BEVFormer) are external prior works, and the paper builds on MapTracker's memory mechanism with attribution. No uniqueness theorem is invoked. The paper honestly reports a limitation (occluded regions are handled poorly, Sec. 5) and an ablation (Table 2) that does not isolate covariance/sampling from deterministic offset refinement; that is an attribution/experimental-design concern, not circularity. A missing ablation of K=1 vs K>1 or deterministic mean at inference would strengthen the probabilistic claim, but the absence does not make the derivation circular. Score 0.

Assumptions & free parameters 3 free parameters · 4 assumptions · 0 invented entities

No new physical entities are introduced. The method rests on a handful of hand-chosen hyperparameters (K, T, memory length) and on domain assumptions about calibration accuracy, unimodal uncertainty, and the validity of the geographic evaluation split.

free parameters (3)

K (number of Gaussian samples per BEV cell)
Hand-chosen sampling count in Pprob (Eq. 6); not stated in the main text, presumably in the supplement. Controls the fidelity of the probabilistic projection.
T (temporal subset size) = 4
Number of historical BEV features selected for temporal fusion, following MapTracker; set by hand.
Memory length for BEV history = 20
Number of past BEV representations stored, following MapTracker's memory mechanism; set by hand.

assumptions (4)

domain assumption Camera intrinsics and extrinsics are known and sufficiently accurate for the static pull mapping
Section 3: 'Assuming known intrinsics and extrinsics, we can define the mapping µpull'. If calibration is poor, the static map is wrong and offsets may not compensate.
domain assumption A single Gaussian per BEV cell adequately models mapping uncertainty
Eq. 6 in Sec. 3.1 defines Pprob sampling from N(µ_i, Σ_i); the paper's limitations section concedes occluded regions are not well handled.
domain assumption The StreamMapNet geographic split is a valid generalization test and all baselines are fairly compared
Section 4.1 relies on the new splits; StreamMapNet results are taken as reported rather than re-run under identical conditions.
standard math Standard deep learning machinery (backprop, differentiable sampling) is available
The sampling in Eq. 6 needs a differentiable path for training; the paper does not explicitly describe the reparameterization or gradient estimator.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Mapping like a Skeptic: Probabilistic BEV Projection for Online HD Mapping." pith.science (2026). https://pith.science/paper/KNZXYBWN

@misc{pith2026250821689,
  author       = {Pith},
  title        = {Pith review of: Mapping like a Skeptic: Probabilistic BEV Projection for Online HD Mapping},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/KNZXYBWN}},
  note         = {Machine review of arXiv:2508.21689}
}

read the original abstract

Constructing high-definition (HD) maps from sensory input requires accurately mapping the road elements in image space to the Bird's Eye View (BEV) space. The precision of this mapping directly impacts the quality of the final vectorized HD map. Existing HD mapping approaches outsource the projection to standard mapping techniques, such as attention-based ones. However, these methods struggle with accuracy due to generalization problems, often hallucinating non-existent road elements. Our key idea is to start with a geometric mapping based on camera parameters and adapt it to the scene to extract relevant map information from camera images. To implement this, we propose a novel probabilistic projection mechanism with confidence scores to (i) refine the mapping to better align with the scene and (ii) filter out irrelevant elements that should not influence HD map generation. In addition, we improve temporal processing by using confidence scores to selectively accumulate reliable information over time. Experiments on new splits of the nuScenes and Argoverse2 datasets demonstrate improved performance over state-of-the-art approaches, indicating better generalization. The improvements are particularly pronounced on nuScenes and in the challenging long perception range. Our code and model checkpoints are available at https://github.com/Fatih-Erdogan/mapping-like-skeptic .

Figures

Figures reproduced from arXiv: 2508.21689 by the authors.

**Figure 1.** Overall Framework. We start by projecting camera features onto the BEV grid using probabilistic projection (detailed in [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗

**Figure 2.** Camera-to-BEV Transformation using Probabilistic Projection. We illustrate the camera-to-BEV transformation for a single camera by dropping the subscript c in the notation. The camera image is first processed by an encoder to extract features F. These features are then processed by two networks, foffset and fconf, to predict the probabilistic projection parameters. Given a distance mask d mask as an additional input… view at source ↗

**Figure 3.** Qualitative Comparison. We qualitatively compare our method to MapTracker [4]. Pedestrian crossings, lane dividers, and road boundaries are shown in blue, red, and green, respectively. Our improvements are highlighted with a purple circle for road boundaries (left), pedestrian crossings (middle), and lane dividers (right) on nuScenes. 4.3 Ablation Study In [PITH_FULL_IMAGE:figures/full_fig_p009_3.png] view at source ↗

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

32 extracted references · 25 canonical work pages

[1]

nuscenes: A multimodal 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 multimodal dataset for autonomous driving. In CVPR, 2020

2020
[2]

End-to-end object detection with transformers

Nicolas Carion, Francisco Massa, Gabriel Synnaeve, Nicolas Usunier, Alexander Kir- illov, and Sergey Zagoruyko. End-to-end object detection with transformers. InECCV, 2020

work page 2020
[3]

PointBeV: A sparse approach for bev predictions

Loick Chambon, Eloi Zablocki, Mickaël Chen, Florent Bartoccioni, Patrick Pérez, and Matthieu Cord. PointBeV: A sparse approach for bev predictions. In CVPR, 2024

work page 2024
[4]

Maptracker: Tracking with strided memory fusion for consistent vector hd mapping

Jiacheng Chen, Yuefan Wu, Jiaqi Tan, Hang Ma, and Yasutaka Furukawa. Maptracker: Tracking with strided memory fusion for consistent vector hd mapping. In ECCV, 2024

work page 2024
[5]

Efficient and robust 2D-to-BEV representation learning via geometry- guided kernel transformer

Shaoyu Chen, Tianheng Cheng, Xinggang Wang, Wenming Meng, Qian Zhang, and Wenyu Liu. Efficient and robust 2D-to-BEV representation learning via geometry- guided kernel transformer. arXiv preprint arXiv:2206.04584, 2022

arXiv 2022
[6]

Mask2map: Vectorized hd map construction using bird’s eye view segmentation masks

Sehwan Choi, Jungho Kim, Hongjae Shin, and Jun Won Choi. Mask2map: Vectorized hd map construction using bird’s eye view segmentation masks. In ECCV, 2024

work page 2024
[7]

Pivotnet: Vectorized pivot learning for end-to-end hd map construction

Wenjie Ding, Limeng Qiao, Xi Qiu, and Chi Zhang. Pivotnet: Vectorized pivot learning for end-to-end hd map construction. In ICCV, 2023

2023
[8]

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

Chunrui Han, Jinrong Yang, Jianjian Sun, Zheng Ge, Runpei Dong, Hongyu Zhou, Weixin Mao, Yuang Peng, and Xiangyu Zhang. Exploring recurrent long-term temporal fusion for multi-view 3D perception. RAL, 9(7):6544–6551, 2024

work page 2024

Show all 32 references

[9]

Simple-bev: What really matters for multi-sensor bev perception? In ICRA, 2023

Adam W Harley, Zhaoyuan Fang, Jie Li, Rares Ambrus, and Katerina Fragkiadaki. Simple-bev: What really matters for multi-sensor bev perception? In ICRA, 2023

2023
[10]

Deep residual learning for image recognition

Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Deep residual learning for image recognition. In CVPR, 2016

2016
[11]

ADMap: Anti-disturbance framework for vectorized hd map construction

Haotian Hu, Fanyi Wang, Yaonong Wang, Laifeng Hu, Jingwei Xu, and Zhiwang Zhang. ADMap: Anti-disturbance framework for vectorized hd map construction. In ECCV, 2024

2024
[12]

Bevpoolv2: A cutting-edge implementation of bevdet toward deployment

Junjie Huang and Guan Huang. Bevpoolv2: A cutting-edge implementation of bevdet toward deployment. arXiv preprint arXiv:2211.17111, 2022

2022 arXiv
[13]

Unveiling the hidden: Online vectorized HD map construction with clip-level token interaction and propagation

Nayeon Kim, Hongje Seong, Daehyun Ji, and Sujin Jang. Unveiling the hidden: Online vectorized HD map construction with clip-level token interaction and propagation. In NeurIPS, 2025

2025
[14]

Hdmapnet: An online hd map con- struction and evaluation framework

Qi Li, Yue Wang, Yilun Wang, and Hang Zhao. Hdmapnet: An online hd map con- struction and evaluation framework. In ICRA, 2022

2022
[15]

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

Zhiqi Li, Wenhai Wang, Hongyang Li, Enze Xie, Chonghao Sima, Tong Lu, Yu Qiao, and Jifeng Dai. Bevformer: Learning bird’s-eye-view representation from multi- camera images via spatiotemporal transformers. In ECCV, 2022. 12 F. ERDOGAN ET AL.: MAPPING LIKE A SKEPTIC

2022
[16]

MapTR: Structured modeling and learning for online vector- ized hd map construction

Bencheng Liao, Shaoyu Chen, Xinggang Wang, Tianheng Cheng, Qian Zhang, Wenyu Liu, and Chang Huang. MapTR: Structured modeling and learning for online vector- ized hd map construction. In ICLR, 2023

2023
[17]

Maptrv2: An end-to-end framework for online vectorized hd map construction

Bencheng Liao, Shaoyu Chen, Yunchi Zhang, Bo Jiang, Qian Zhang, Wenyu Liu, Chang Huang, and Xinggang Wang. Maptrv2: An end-to-end framework for online vectorized hd map construction. IJCV, pages 1–23, 2024

2024
[18]

Mgmap: Mask-guided learning for online vectorized hd map construction

Xiaolu Liu, Song Wang, Wentong Li, Ruizi Yang, Junbo Chen, and Jianke Zhu. Mgmap: Mask-guided learning for online vectorized hd map construction. In CVPR, 2024

2024
[19]

Vectormapnet: End-to-end vectorized hd map learning

Yicheng Liu, Tianyuan Yuan, Yue Wang, Yilun Wang, and Hang Zhao. Vectormapnet: End-to-end vectorized hd map learning. In ICLR, 2023

2023
[20]

Leveraging en- hanced queries of point sets for vectorized map construction

Zihao Liu, Xiaoyu Zhang, Guangwei Liu, Ji Zhao, and Ningyi Xu. Leveraging en- hanced queries of point sets for vectorized map construction. In ECCV, 2024

2024
[21]

PrevPredMap: Exploring temporal modeling with previous predictions for on- line vectorized hd map construction

Nan Peng, Xun Zhou, Mingming Wang, Xiaojun Yang, Songming Chen, and Guisong Chen. PrevPredMap: Exploring temporal modeling with previous predictions for on- line vectorized hd map construction. arXiv preprint arXiv:2407.17378, 2024

2024 arXiv
[22]

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, 2020

2020
[23]

End-to-end vectorized hd-map construction with piecewise bezier curve

Limeng Qiao, Wenjie Ding, Xi Qiu, and Chi Zhang. End-to-end vectorized hd-map construction with piecewise bezier curve. In CVPR, 2023

2023
[24]

MemFu- sionMap: Working memory fusion for online vectorized hd map construction

Jingyu Song, Xudong Chen, Liupei Lu, Jie Li, and Katherine A Skinner. MemFu- sionMap: Working memory fusion for online vectorized hd map construction. arXiv preprint arXiv:2409.18737, 2024

2024 arXiv
[25]

Exploring object-centric temporal modeling for efficient multi-view 3d object detection

Shihao Wang, Yingfei Liu, Tiancai Wang, Ying Li, and Xiangyu Zhang. Exploring object-centric temporal modeling for efficient multi-view 3d object detection. InICCV, 2023

2023
[26]

Stream query denoising for vec- torized HD-map construction

Shuo Wang, Fan Jia, Weixin Mao, Yingfei Liu, Yucheng Zhao, Zehui Chen, Tiancai Wang, Chi Zhang, Xiangyu Zhang, and Feng Zhao. Stream query denoising for vec- torized HD-map construction. In ECCV, 2024

2024
[27]

Argoverse 2: Next generation datasets for self-driving perception and forecasting

Benjamin Wilson, William Qi, Tanmay Agarwal, John Lambert, Jagjeet Singh, Sid- dhesh Khandelwal, Bowen Pan, Ratnesh Kumar, Andrew Hartnett, Jhony Kaesemodel Pontes, et al. Argoverse 2: Next generation datasets for self-driving perception and forecasting. arXiv preprint arXiv:2...

2023 arXiv
[28]

Wong, and Hengshuang Zhao

Zhenhua Xu, Kwan-Yee K. Wong, and Hengshuang Zhao. InsMapper: Exploring inner- instance information for vectorized hd mapping. In ECCV, 2024

2024
[29]

Streammapnet: Streaming mapping network for vectorized online hd map construction

Tianyuan Yuan, Yicheng Liu, Yue Wang, Yilun Wang, and Hang Zhao. Streammapnet: Streaming mapping network for vectorized online hd map construction. In WACV, 2024. F. ERDOGAN ET AL.: MAPPING LIKE A SKEPTIC 13

2024
[30]

Online vectorized HD map construction using geometry

Zhixin Zhang, Yiyuan Zhang, Xiaohan Ding, Fusheng Jin, and Xiangyu Yue. Online vectorized HD map construction using geometry. In ECCV, 2024

2024
[31]

Cross-view transformers for real-time map-view semantic segmentation

Brady Zhou and Philipp Krähenbühl. Cross-view transformers for real-time map-view semantic segmentation. In CVPR, 2022

2022
[32]

Himap: Hybrid representation learning for end-to-end vectorized hd map construction

Yi Zhou, Hui Zhang, Jiaqian Yu, Yifan Yang, Sangil Jung, Seung-In Park, and Byun- gIn Yoo. Himap: Hybrid representation learning for end-to-end vectorized hd map construction. In CVPR, 2024

2024

Pith tools

Reviewed August 5, 2026 · model on record in the stance chip above.

[1] [1]

nuscenes: A multimodal 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 multimodal dataset for autonomous driving. In CVPR, 2020

2020

[2] [2]

End-to-end object detection with transformers

Nicolas Carion, Francisco Massa, Gabriel Synnaeve, Nicolas Usunier, Alexander Kir- illov, and Sergey Zagoruyko. End-to-end object detection with transformers. InECCV, 2020

work page 2020

[3] [3]

PointBeV: A sparse approach for bev predictions

Loick Chambon, Eloi Zablocki, Mickaël Chen, Florent Bartoccioni, Patrick Pérez, and Matthieu Cord. PointBeV: A sparse approach for bev predictions. In CVPR, 2024

work page 2024

[4] [4]

Maptracker: Tracking with strided memory fusion for consistent vector hd mapping

Jiacheng Chen, Yuefan Wu, Jiaqi Tan, Hang Ma, and Yasutaka Furukawa. Maptracker: Tracking with strided memory fusion for consistent vector hd mapping. In ECCV, 2024

work page 2024

[5] [5]

Efficient and robust 2D-to-BEV representation learning via geometry- guided kernel transformer

Shaoyu Chen, Tianheng Cheng, Xinggang Wang, Wenming Meng, Qian Zhang, and Wenyu Liu. Efficient and robust 2D-to-BEV representation learning via geometry- guided kernel transformer. arXiv preprint arXiv:2206.04584, 2022

arXiv 2022

[6] [6]

Mask2map: Vectorized hd map construction using bird’s eye view segmentation masks

Sehwan Choi, Jungho Kim, Hongjae Shin, and Jun Won Choi. Mask2map: Vectorized hd map construction using bird’s eye view segmentation masks. In ECCV, 2024

work page 2024

[7] [7]

Pivotnet: Vectorized pivot learning for end-to-end hd map construction

Wenjie Ding, Limeng Qiao, Xi Qiu, and Chi Zhang. Pivotnet: Vectorized pivot learning for end-to-end hd map construction. In ICCV, 2023

2023

[8] [8]

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

Chunrui Han, Jinrong Yang, Jianjian Sun, Zheng Ge, Runpei Dong, Hongyu Zhou, Weixin Mao, Yuang Peng, and Xiangyu Zhang. Exploring recurrent long-term temporal fusion for multi-view 3D perception. RAL, 9(7):6544–6551, 2024

work page 2024