Pith. sign in

REVIEW 3 major objections 5 minor 48 references

Efficient Masked AutoEncoder for Video Object Counting and A Large-Scale Benchmark

T0 review · 3 major / 5 minor · reviewed 2026-08-12 · deepseek-v4-flash

Pith's one-line read A masked autoencoder that treats density maps as a second modality sets new accuracy marks on four video counting benchmarks while introducing a drone-captured bird-counting dataset.

desk verdict Useful new video bird benchmark and a coherent masked-autoencoder counting framework, but the SOTA numbers rest on baselines that don't share the same density-map protocol. read the letter →

arxiv 2411.13056 v2 pith:U5ZASQWC submitted 2024-11-20 cs.CV

classification cs.CV
keywords videoobjectcountingmaskedautoencoderdensitymapregressioncrowdbirddatasetopticalflowtemporalfusionself-supervisedrepresentationlearningspatialadaptivemasking
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

Video frames of crowds and animals are often sparse, with most pixels belonging to empty background, which makes density-counting models over- or under-count. The paper proposes E-MAC, a masked-autoencoder framework that treats the ground-truth density map as a second input modality, reconstructs masked density patches, and uses the density distribution to decide which image patches to keep, so the model spends its capacity on foreground objects. An optical-flow-based fusion module then warps the previous frame's predicted density map into the current frame and adds residual corrections. The authors report state-of-the-art mean absolute error on all four tested benchmarks (Mall, FDST, VSCrowd, and their new DroneBird), and they introduce DroneBird, a drone-captured video dataset of migratory birds. If the results hold, the framework gives video counting a reusable recipe: let the density map itself supervise representation learning rather than only serving as the regression target.

What carries the argument

The load-bearing object is the density map used as an auxiliary modality inside a masked autoencoder: instead of only being the regression target, the density map also selects which image tokens survive masking and provides the reconstruction signal. This transforms counting into a self-supervised cross-modal reconstruction problem. The supporting mechanisms are SAM (spatial adaptive masking, which sorts image tokens by the object count in their corresponding density patch and keeps the top foreground tokens with probability 1-P) and TCF (temporal collaborative fusion, which warps the previous frame's density prediction to the current frame using optical flow and adds a cross-attention residual).

What would settle it

Run each video-counting baseline on Mall, FDST, VSCrowd, and DroneBird under the paper's exact protocol (fixed Gaussian sigma=6, same input sizes, same standardization), then recompute MAE/RMSE; if the best baselines match E-MAC within noise, the claimed state-of-the-art advantage would not survive.

Watch

Extended reading notes

Core claim

The paper's central claim is that density-embedded masked modeling solves the foreground-background imbalance that hurts video object counting. In DEMO, image patches and density-map patches are tokenized together; a density-guided Spatial Adaptive Masking (SAM) keeps image tokens whose corresponding density patches contain many objects (with a small background-retention probability), while density-map tokens are randomly masked. The decoder reconstructs the full density map from the surviving tokens, so counting is learned as a self-supervised reconstruction task. A Temporal Collaborative Fusion (TCF) module computes optical flow between adjacent frames, warps the previous predicted density map, and uses cross-attention to produce residual corrections that are added to the current prediction. The authors report the lowest MAE and RMSE on Mall, FDST, and VSCrowd, and on the new DroneBird dataset (e.g., MAE 1.35/1.29/6.0/38.72 respectively), with the largest gains on DroneBird where the MAE is claimed 58% below STGN.

Load-bearing premise

The comparisons with prior methods assume those methods were evaluated under the same density-map generation and evaluation protocol (a fixed Gaussian kernel with sigma=6 in this paper); if the published baseline numbers used different kernels or preprocessing, the reported performance gaps may partly reflect protocol differences rather than counting accuracy.

Editorial extensions

If this is right

  • Counting accuracy improves by using the density map as a second modality during masked modeling, not just as the training target; the ablation shows DEMO alone cuts FDST MAE by roughly 27% over the baseline.
  • Spatial adaptive masking concentrates computation on foreground patches, reducing redundant background learning; it contributes the largest single ablation gain (about 32% MAE on FDST).
  • Temporal collaborative fusion with optical flow combines adjacent-frame density predictions and lowers error on video benchmarks (about 5 to 16 percent on FDST in the ablation).
  • The DroneBird dataset supports evaluation on small, fast-moving, sparsely distributed targets in natural scenes; transferring a crowd-trained model to DroneBird fails (MAE 183.31 vs. 38.72 when trained on target), showing the domain gap.

Reading between the lines

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

  • The same density-embedded masked modeling recipe likely transfers to other sparse-object dense-regression tasks (cell counting, vehicle counting from drones), since the underlying mechanism of density-guided token selection inside masked reconstruction does not assume human targets.
  • Because the mask ratio and background-retention probability (P=0.2) are tuned on FDST, applying E-MAC to new domains will probably require re-tuning these two hyperparameters; the paper's own experiments show both have non-monotonic effects.
  • A natural next test is whether the density map used for masking can be replaced by a coarse predicted density at inference, which would relax the need for point annotations during deployment and potentially enable self-supervised adaptation to unseen scenes.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 5 minor

Summary. This paper proposes an Efficient Masked Autoencoder Counting (E-MAC) framework for video object counting. The method combines three components: density-embedded masked modeling (DEMO), which treats the density map as an auxiliary modality during masked autoencoding; spatial adaptive masking (SAM), which prunes background tokens based on the density map; and temporal collaborative fusion (TCF), which uses optical flow to warp the previous frame's predicted density map and compute temporal residuals. The authors also introduce DroneBird, a large drone-captured video dataset with 3.69M bird annotations and trajectory annotations. Experiments on Mall, FDST, VSCrowd and DroneBird report state-of-the-art MAE/RMSE, with the best MAE of 1.35, 1.29, 6.0 and 38.72, respectively.

Significance. The dataset contribution is substantial: DroneBird appears to be the first drone-view video bird counting dataset with both point and trajectory annotations, containing 3.69M annotations, and the authors make code and data available. The methodological idea of using the density map as a second modality in masked autoencoding is novel and well motivated by the fore-background imbalance problem. The ablation study is systematic and includes a random-seed robustness check on FDST. If the reported gains survive protocol-matched evaluation, the work would be a meaningful advance for video object counting. However, the current empirical evidence for state-of-the-art performance is weakened by the evaluation protocol mismatch described below.

major comments (3)
  1. [Section 5.1, Table 1] Section 5.1 states that 'We use a fixed Gaussian kernel (sigma = 6) to generate the ground-truth density map on these datasets,' yet the baseline MAE/RMSE values for Mall, FDST and VSCrowd in Table 1 are quoted from earlier papers that used their own density-map generation protocols (e.g., geometry-adaptive kernels in CSRNet/MAN). Because MAE and RMSE are computed against ground-truth density maps, changing the kernel changes the regression target and the metric even for identical predictions. The margins on Mall (1.35 vs 1.41) and FDST (1.29 vs 1.38) are small enough that a kernel/protocol shift could plausibly reverse them. Please retrain or re-evaluate all baselines under the exact sigma=6 protocol (and the same input resolutions and preprocessing), or present the comparisons only for protocol-matched runs.
  2. [Section 5.4, Tables 2-3] The hyperparameters P, the mask ratio, and the loss weights {lambda_1..lambda_4} are tuned via experiments on FDST in Section 5.4 and Figure 5, after which FDST is reported as a headline result in Table 1. If these experiments used the FDST test frames, the reported 1.29/1.69 is a test-set-tuned estimate and is not an unbiased measure of generalization. Please clarify whether the tuning was done on a held-out validation split; if not, report results with hyperparameters fixed a priori or tuned only on validation data.
  3. [Table 1, Section 5.2] The paper does not state which baseline numbers were produced by the authors' own re-runs and which were transcribed from prior publications. The DroneBird rows presumably come from the authors' implementations, whereas Mall/FDST/VSCrowd rows appear to be copied from earlier papers with heterogeneous protocols. This mixed provenance makes it impossible for the reader to separate algorithmic gains from evaluation-configuration gains. Please provide a source statement for every number in Table 1, along with the exact evaluation code and, where feasible, standard deviations over at least three runs.
minor comments (5)
  1. [Table 12] The RMSE value for the proposed method is written as '42,92'; it should be '42.92'.
  2. [Figure 5] Subfigures (c)-(f) lack axis labels identifying which loss weight is being swept, and the caption does not state the fixed values of the other weights; please label each subfigure and state the fixed settings.
  3. [Algorithm 2, Section 4.3] The number of retained density tokens N_ret^D and its relationship to N_ret^I (including how the Dirichlet distribution determines them) are mentioned in the appendix but not defined in Section 4.3; please define both quantities in the main text.
  4. [Section 5.2] The dataset name is spelled 'VScrowd' in the text (e.g., 'VScrowd collected more videos') but 'VSCrowd' elsewhere; please unify the spelling.
  5. [Equation (6)] Equation (6) introduces lambda_1..lambda_4 without mapping them explicitly to L_fuse, L_cur, L_opt, and L_TV; a one-sentence mapping would improve readability.

Circularity Check

1 steps flagged · score 6.0 of 10

FDST headline result is partially circular: hyperparameters and components were selected on the FDST test metric, then the FDST SOTA is reported as a prediction; the other three benchmarks remain independent.

  1. fitted input called prediction [Section 5.4 (Impact of Background Retention Probability and Hyperparameter Analysis), built on Section 5.3 ablations on FDST; reported in Section 5.2 and Table 1.]
    "The horizontal axis indicates the probability of sorting the tokens in ascending order. We notice that the curve shows a clear downward rebound trend, and the quantitative metrics show a decline of different degrees in both four experiments compared to Exp.V. We finally choose 0.2 as the default probability in our experiments."

    The paper selects the background retention probability P, the mask ratio, the loss weights lambda1-lambda4, and the inclusion of DEMO/SAM/TCF by running ablations and hyperparameter analyses on the FDST dataset, using the same MAE/RMSE that later appears as the FDST headline result in Table 1 and Section 5.2 ('our method achieves the best MAE and RMSE, decreasing the two metrics of 7% compared to the runner-up method STGN'). Thus the FDST number is not an independent prediction but a configuration-selected outcome: the reported metric on FDST is the criterion that was optimized during model selection, so that particular SOTA claim is statistically forced rather than a genuine out-of-sample result.

full rationale

This is an empirical neural-network paper, so there is no equation-level derivation chain whose conclusion equals its premise. The E-MAC components (DEMO, SAM, TCF) are trained and evaluated on data splits, and the DroneBird comparisons appear protocol-matched because the authors ran the baselines on their own dataset. The one circular step I can exhibit by precise quotation is the FDST pipeline: all ablations and hyperparameter choices are performed on the FDST dataset using MAE/RMSE, and the same FDST MAE/RMSE is then reported as the SOTA result. That matches the fitted-input-called-prediction pattern for FDST. The possible mismatch in density-map generation protocols (fixed sigma=6 here versus protocols in the cited baseline papers) is a validity confound for Table 1 comparisons, but it is not circularity, so it is not scored as a circular step. No load-bearing self-citation chain or definitional equivalence was found. Score 6 reflects one genuine partial circularity; the other three benchmark claims retain independent content.

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

The model relies on a fixed density-map generation kernel, a pretrained optical flow network, a train-inference masking mismatch, and comparability of external baseline numbers. These are domain assumptions rather than fitted physical parameters. The free parameters are the typical hyperparameters of the training setup, all tuned on a single dataset (FDST).

free parameters (4)
  • sigma (Gaussian kernel size) = 6
    Used to generate ground-truth density maps from point annotations on all datasets. Chosen by hand, not tuned per dataset, and may not match the density-map protocols used by baseline papers.
  • mask ratio = 0.72
    Fraction of tokens masked in DEMO. Selected via experiments on FDST (Sec 5.4).
  • background retention probability P = 0.2
    Probability of retaining background tokens in Spatial Adaptive Masking. Selected via experiments on FDST (Sec 5.4).
  • loss weights lambda1..lambda4 = 10, 10, 1, 20
    Weights of Lfuse, Lcur, Lopt, LTV. Chosen based on hyperparameter analysis on FDST (Sec 5.4).
assumptions (4)
  • domain assumption Ground-truth density maps are reliable supervision for counting accuracy.
    The entire training and evaluation rely on density maps generated from point annotations with a fixed Gaussian kernel, a common but nontrivial modeling choice that affects reconstruction targets and metrics.
  • domain assumption Pretrained optical flow (PWCNet) provides accurate inter-frame alignment for the counting domain, including small fast-moving birds.
    Temporal Collaborative Fusion warps predicted density maps using optical flow computed on frames; if optical flow is unreliable on tiny targets, the fusion could add noise rather than signal (Sec 4.1).
  • domain assumption The model generalizes from training with visible density tokens to inference with all density tokens masked.
    Training keeps N_ret^D density tokens, while inference sets N_ret^D=0; the authors assume the decoder can reconstruct complete density maps from image tokens alone (Sec 4.3).
  • domain assumption Baseline numbers cited from prior papers are comparable to E-MAC's protocol.
    No baselines are retrained under the same density-map kernel, image size, and evaluation split; comparability is assumed implicitly in Tables 1, 6-8.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Efficient Masked AutoEncoder for Video Object Counting and A Large-Scale Benchmark." pith.science (2026). https://pith.science/paper/U5ZASQWC

@misc{pith2026241113056,
  author       = {Pith},
  title        = {Pith review of: Efficient Masked AutoEncoder for Video Object Counting and A Large-Scale Benchmark},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/U5ZASQWC}},
  note         = {Machine review of arXiv:2411.13056}
}
abstract

The dynamic imbalance of the fore-background is a major challenge in video object counting, which is usually caused by the sparsity of target objects. This remains understudied in existing works and often leads to severe under-/over-prediction errors. To tackle this issue in video object counting, we propose a density-embedded Efficient Masked Autoencoder Counting (E-MAC) framework in this paper. To empower the model's representation ability on density regression, we develop a new $\mathtt{D}$ensity-$\mathtt{E}$mbedded $\mathtt{M}$asked m$\mathtt{O}$deling ($\mathtt{DEMO}$) method, which first takes the density map as an auxiliary modality to perform multimodal self-representation learning for image and density map. Although $\mathtt{DEMO}$ contributes to effective cross-modal regression guidance, it also brings in redundant background information, making it difficult to focus on the foreground regions. To handle this dilemma, we propose an efficient spatial adaptive masking derived from density maps to boost efficiency. Meanwhile, we employ an optical flow-based temporal collaborative fusion strategy to effectively capture the dynamic variations across frames, aligning features to derive multi-frame density residuals. The counting accuracy of the current frame is boosted by harnessing the information from adjacent frames. In addition, considering that most existing datasets are limited to human-centric scenarios, we first propose a large video bird counting dataset, DroneBird, in natural scenarios for migratory bird protection. Extensive experiments on three crowd datasets and our \textit{DroneBird} validate our superiority against the counterparts. The code and dataset are available.

Figures

Figures reproduced from arXiv: 2411.13056 by the authors.

Figure 1
Figure 1. The chord diagram illustrates the associations between various attributes of our proposed [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. An overview of our E-MAC. For the temporal collaborative fusion, we use optical flow [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Visualized comparisons on the FDST dataset and the Mall dataset. [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗
Figures from the paper (9 more)
Figure 4
Figure 4. Figure 4: Our method achieves accurate counting results under low-light or long-distance dense con [PITH_FULL_IMAGE:figures/full_fig_p008_4.png]
Figure 5
Figure 5. Figure 5: Hyperparameter analysis of background retention probability, mask ratio, and loss weights. [PITH_FULL_IMAGE:figures/full_fig_p010_5.png]
Figure 6
Figure 6. Figure 6: Visualization of partial examples of DroneBird. [PITH_FULL_IMAGE:figures/full_fig_p014_6.png]
Figure 7
Figure 7. Figure 7: Visualization of pixel occupied by the target in DroneBird and existing crowd dataset. [PITH_FULL_IMAGE:figures/full_fig_p014_7.png]
Figure 8
Figure 8. Figure 8: The detailed process of SAM. For ease of expression, we crop the image and the density [PITH_FULL_IMAGE:figures/full_fig_p017_8.png]
Figure 9
Figure 9. Figure 9: Visualization of the output and intermediate variables in the Fusion Module. [PITH_FULL_IMAGE:figures/full_fig_p020_9.png]
Figure 10
Figure 10. Figure 10: Visualization of the predicted density map w/ and w/o [PITH_FULL_IMAGE:figures/full_fig_p020_10.png]
Figure 11
Figure 11. Figure 11: Visualization of the predicted results of our key components ( [PITH_FULL_IMAGE:figures/full_fig_p021_11.png]
Figure 12
Figure 12. Figure 12: Density distribution. The bar graph portion (blue) represents the number of patches corresponding to the crowd density. The line graph portion (red) represents the percentage of the number of patches whose density is less than the current density. Statistical result i…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

48 extracted references · 28 canonical work pages

  1. [1]

    write newline

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

  2. [2]

    Arteta, V

    C. Arteta, V. Lempitsky, and A. Zisserman. Counting in the wild. In European Conference on Computer Vision, 2016

  3. [3]

    A spatio-temporal attentive network for video-based crowd counting

    Marco Avvenuti, Marco Bongiovanni, Luca Ciampi, Fabrizio Falchi, Claudio Gennaro, and Nicola Messina. A spatio-temporal attentive network for video-based crowd counting. In 2022 IEEE Symposium on Computers and Communications (ISCC), pp.\ 1--6, 2022. doi:10.1109/ISCC55528.2022.9913019

  4. [4]

    MultiMAE : Multi-modal multi-task masked autoencoders

    Roman Bachmann, David Mizrahi, Andrei Atanov, and Amir Zamir. MultiMAE : Multi-modal multi-task masked autoencoders. European Conference on Computer Vision, 2022

  5. [5]

    Haoyue Bai and S. H. Gary Chan. Motion-guided non-local spatial-temporal network for video crowd counting, 2021

  6. [6]

    BE it: BERT pre-training of image transformers

    Hangbo Bao, Li Dong, Songhao Piao, and Furu Wei. BE it: BERT pre-training of image transformers. In International Conference on Learning Representations, 2022. URL https://openreview.net/forum?id=p-BhZSz59o4

  7. [7]

    Generative pretraining from pixels

    Mark Chen, Alec Radford, Rewon Child, Jeffrey Wu, Heewoo Jun, David Luan, and Ilya Sutskever. Generative pretraining from pixels. In Hal Daumé III and Aarti Singh (eds.), Proceedings of the 37th International Conference on Machine Learning, volume 119 of Proceedings of Machine Learning Research, pp.\ 1691--1703. PMLR, 13--18 Jul 2020. URL https://proceedi...

  8. [8]

    BERT : Pre-training of deep bidirectional transformers for language understanding

    Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. BERT : Pre-training of deep bidirectional transformers for language understanding. In Jill Burstein, Christy Doran, and Thamar Solorio (eds.), Proceedings of the 2019 Conference of the North A merican Chapter of the Association for Computational Linguistics: Human Language Technologies, Vol...

Show all 48 references
  1. [9]

    An image is worth 16x16 words: Transformers for image recognition at scale

    Alexey Dosovitskiy, Lucas Beyer, Alexander Kolesnikov, Dirk Weissenborn, Xiaohua Zhai, Thomas Unterthiner, Mostafa Dehghani, Matthias Minderer, Georg Heigold, Sylvain Gelly, et al. An image is worth 16x16 words: Transformers for image recognition at scale. arXiv preprint arXiv...

  2. [10]

    Redesigning multi-scale neural network for crowd counting

    Zhipeng Du, Miaojing Shi, Jiankang Deng, and Stefanos Zafeiriou. Redesigning multi-scale neural network for crowd counting. IEEE Transactions on Image Processing, 2023

  3. [11]

    Locality-constrained spatial transformer network for video crowd counting

    Yanyan Fang, Biyun Zhan, Wandi Cai, Shenghua Gao, and Bo Hu. Locality-constrained spatial transformer network for video crowd counting. arXiv preprint arXiv:1907.07911, 2019

  4. [12]

    Multi-level feature fusion based locality-constrained spatial transformer network for video crowd counting

    Yanyan Fang, Shenghua Gao, Jing Li, Weixin Luo, Linfang He, and Bo Hu. Multi-level feature fusion based locality-constrained spatial transformer network for video crowd counting. Neurocomputing, 392: 0 98--107, 2020. ISSN 0925-2312. doi:https://doi.org/10.1016/j.neucom.2020.01...

  5. [13]

    Masked autoencoders are scalable vision learners

    Kaiming He, Xinlei Chen, Saining Xie, Yanghao Li, Piotr Doll \'a r, and Ross Girshick. Masked autoencoders are scalable vision learners. arXiv:2111.06377, 2021

  6. [14]

    Video-based crowd counting using a multi-scale optical flow pyramid network

    Mohammad Asiful Hossain, Kevin Cannons, Daesik Jang, Fabio Cuzzolin, and Zhan Xu. Video-based crowd counting using a multi-scale optical flow pyramid network. In Proceedings of the Asian Conference on Computer Vision (ACCV), November 2020

  7. [15]

    Frame-recurrent video crowd counting

    Yi Hou, Shanghang Zhang, Rui Ma, Huizhu Jia, and Xiaodong Xie. Frame-recurrent video crowd counting. IEEE Transactions on Circuits and Systems for Video Technology, 33 0 (9): 0 5186--5199, 2023. doi:10.1109/TCSVT.2023.3250946

  8. [16]

    Clip-count: Towards text-guided zero-shot object counting

    Ruixiang Jiang, Lingbo Liu, and Changwen Chen. Clip-count: Towards text-guided zero-shot object counting. In Proceedings of the 31st ACM International Conference on Multimedia, MM '23, pp.\ 4535–4545, New York, NY, USA, 2023. Association for Computing Machinery. ISBN 979840070...

  9. [17]

    Vlcounter: Text-aware visual representation for zero-shot object counting

    Seunggu Kang, WonJun Moon, Euiyeon Kim, and Jae-Pil Heo. Vlcounter: Text-aware visual representation for zero-shot object counting. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 38, pp.\ 2714--2722, 2024

  10. [18]

    Video crowd localization with multifocus gaussian neighborhood attention and a large-scale benchmark

    Haopeng Li, Lingbo Liu, Kunlin Yang, Shinan Liu, Junyu Gao, Bin Zhao, Rui Zhang, and Jun Hou. Video crowd localization with multifocus gaussian neighborhood attention and a large-scale benchmark. IEEE Transactions on Image Processing, 31: 0 6032--6047, 2022. doi:10.1109/TIP.20...

  11. [19]

    Csrnet: Dilated convolutional neural networks for understanding the highly congested scenes

    Yuhong Li, Xiaofan Zhang, and Deming Chen. Csrnet: Dilated convolutional neural networks for understanding the highly congested scenes. In 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp.\ 1091--1100, 2018. doi:10.1109/CVPR.2018.00120

  12. [20]

    Transcrowd: weakly-supervised crowd counting with transformers

    Dingkang Liang, Xiwu Chen, Wei Xu, Yu Zhou, and Xiang Bai. Transcrowd: weakly-supervised crowd counting with transformers. Science China Information Sciences, 65 0 (6): 0 1--14, 2022

  13. [21]

    Boosting crowd counting via multifaceted attention

    Hui Lin, Zhiheng Ma, Rongrong Ji, Yaowei Wang, and Xiaopeng Hong. Boosting crowd counting via multifaceted attention. In CVPR, 2022

  14. [22]

    Gramformer: Learning crowd counting via graph-modulated transformer, 2024

    Hui Lin, Zhiheng Ma, Xiaopeng Hong, Qinnan Shangguan, and Deyu Meng. Gramformer: Learning crowd counting via graph-modulated transformer, 2024. URL https://arxiv.org/abs/2401.03870

  15. [23]

    Point-query quadtree for crowd counting, localization, and more

    Chengxin Liu, Hao Lu, Zhiguo Cao, and Tongliang Liu. Point-query quadtree for crowd counting, localization, and more. In Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 2023

  16. [24]

    Context-aware crowd counting

    Weizhe Liu, Mathieu Salzmann, and Pascal Fua. Context-aware crowd counting. In The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), June 2019

  17. [25]

    Estimating people flows to better count them in crowded scenes

    Weizhe Liu, Mathieu Salzmann, and Pascal Fua. Estimating people flows to better count them in crowded scenes. In The European Conference on Computer Vision (ECCV), August 2020

  18. [26]

    From semi-supervised to transfer counting of crowds

    Chen Change Loy, Shaogang Gong, and Tao Xiang. From semi-supervised to transfer counting of crowds. In 2013 IEEE International Conference on Computer Vision, pp.\ 2256--2263, 2013. doi:10.1109/ICCV.2013.270

  19. [27]

    Bayesian loss for crowd count estimation with point supervision

    Zhiheng Ma, Xing Wei, Xiaopeng Hong, and Yihong Gong. Bayesian loss for crowd count estimation with point supervision. In Proceedings of the IEEE International Conference on Computer Vision, pp.\ 6142--6151, 2019

  20. [28]

    Phnet: Parasite-host network for video crowd counting

    Shiqiao Meng, Jiajie Li, Weiwei Guo, Lai Ye, and Jinfeng Jiang. Phnet: Parasite-host network for video crowd counting. In 2020 25th International Conference on Pattern Recognition (ICPR), pp.\ 1956--1963, 2021. doi:10.1109/ICPR48806.2021.9412792

  21. [29]

    Rudin, Stanley Osher, and Emad Fatemi

    Leonid I. Rudin, Stanley Osher, and Emad Fatemi. Nonlinear total variation based noise removal algorithms. Physica D: Nonlinear Phenomena, 60 0 (1): 0 259--268, 1992. ISSN 0167-2789. doi:https://doi.org/10.1016/0167-2789(92)90242-F. URL https://www.sciencedirect.com/science/ar...

  22. [30]

    Convolutional lstm network: a machine learning approach for precipitation nowcasting

    Xingjian Shi, Zhourong Chen, Hao Wang, Dit-Yan Yeung, Wai-kin Wong, and Wang-chun Woo. Convolutional lstm network: a machine learning approach for precipitation nowcasting. In Proceedings of the 28th International Conference on Neural Information Processing Systems - Volume 1,...

  23. [31]

    Crowd counting in the frequency domain

    Weibo Shu, Jia Wan, Kay Chen Tan, Sam Kwong, and Antoni B Chan. Crowd counting in the frequency domain. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp.\ 19618--19627, 2022

  24. [32]

    PWC-Net : CNNs for optical flow using pyramid, warping, and cost volume

    Deqing Sun, Xiaodong Yang, Ming-Yu Liu, and Jan Kautz. PWC-Net : CNNs for optical flow using pyramid, warping, and cost volume. In 2018 IEEE Conference on Computer Vision and Pattern Recognition, 2018

  25. [33]

    Cctrans: Simplifying and improving crowd counting with transformer

    Ye Tian, Xiangxiang Chu, and Hongpeng Wang. Cctrans: Simplifying and improving crowd counting with transformer. arXiv preprint arXiv:2109.14483, 2021

  26. [34]

    Video MAE : Masked autoencoders are data-efficient learners for self-supervised video pre-training

    Zhan Tong, Yibing Song, Jue Wang, and Limin Wang. Video MAE : Masked autoencoders are data-efficient learners for self-supervised video pre-training. In Advances in Neural Information Processing Systems, 2022

  27. [35]

    Attention is all you need

    Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, ukasz Kaiser, and Illia Polosukhin. Attention is all you need. Advances in neural information processing systems, 30, 2017

  28. [36]

    Extracting and composing robust features with denoising autoencoders

    Pascal Vincent, Hugo Larochelle, Yoshua Bengio, and Pierre-Antoine Manzagol. Extracting and composing robust features with denoising autoencoders. In Proceedings of the 25th International Conference on Machine Learning, ICML '08, pp.\ 1096–1103, New York, NY, USA, 2008. Associ...

  29. [37]

    Bird-count: a multi-modality benchmark and system for bird population counting in the wild

    Hongchang Wang, Huaxiang Lu, Huimin Guo, Haifang Jian, Chuang Gan, and Wu Liu. Bird-count: a multi-modality benchmark and system for bird population counting in the wild. Multimedia Tools and Applications, Apr 2023. ISSN 1573-7721. doi:10.1007/s11042-023-14833-z. URL https://d...

  30. [38]

    Fast video crowd counting with a temporal aware network

    Xingjiao Wu, Baohan Xu, Yingbin Zheng, Hao Ye, Jing Yang, and Liang He. Fast video crowd counting with a temporal aware network. Neurocomputing, 403: 0 13--20, 2020

  31. [39]

    Spatial-temporal graph network for video crowd counting

    Zhe Wu, Xinfeng Zhang, Geng Tian, Yaowei Wang, and Qingming Huang. Spatial-temporal graph network for video crowd counting. IEEE Transactions on Circuits and Systems for Video Technology, 33 0 (1): 0 228--241, 2023. doi:10.1109/TCSVT.2022.3187194

  32. [40]

    Spatiotemporal modeling for crowd counting in videos

    Feng Xiong, Xingjian Shi, and Dit-Yan Yeung. Spatiotemporal modeling for crowd counting in videos. In 2017 IEEE International Conference on Computer Vision (ICCV), pp.\ 5161--5169, 2017. doi:10.1109/ICCV.2017.551

  33. [41]

    Reverse perspective network for perspective-aware object counting

    Yifan Yang, Guorong Li, Zhe Wu, Li Su, Qingming Huang, and Nicu Sebe. Reverse perspective network for perspective-aware object counting. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp.\ 4374--4383, 2020

  34. [42]

    Single-image crowd counting via multi-column convolutional neural network

    Yingying Zhang, Desen Zhou, Siqin Chen, Shenghua Gao, and Yi Ma. Single-image crowd counting via multi-column convolutional neural network. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), June 2016

  35. [43]

    Locality-aware crowd counting

    Joey Tianyi Zhou, Le Zhang, Jiawei Du, Xi Peng, Zhiwen Fang, Zhe Xiao, and Hongyuan Zhu. Locality-aware crowd counting. IEEE Transactions on Pattern Analysis and Machine Intelligence, 44 0 (7): 0 3602--3613, 2022. doi:10.1109/TPAMI.2021.3056518

  36. [44]

    Graph regularized flow attention network for video animal counting from drones

    Pengfei Zhu, Tao Peng, Dawei Du, Hongtao Yu, Libo Zhang, and Qinghua Hu. Graph regularized flow attention network for video animal counting from drones. IEEE Transactions on Image Processing, 30: 0 5339--5351, 2021. doi:10.1109/TIP.2021.3082297

  37. [45]

    Enhanced 3d convolutional networks for crowd counting

    Zhikang Zou, Huiliang Shao, Xiaoye Qu, Wei Wei, and Pan Zhou. Enhanced 3d convolutional networks for crowd counting. arXiv preprint arXiv:1908.04121, 2019

  38. [46]

    @esa (Ref

    \@ifxundefined[1] #1\@undefined \@firstoftwo \@secondoftwo \@ifnum[1] #1 \@firstoftwo \@secondoftwo \@ifx[1] #1 \@firstoftwo \@secondoftwo [2] @ #1 \@temptokena #2 #1 @ \@temptokena \@ifclassloaded agu2001 natbib The agu2001 class already includes natbib coding, so you should ...

  39. [47]

    \@lbibitem[] @bibitem@first@sw\@secondoftwo \@lbibitem[#1]#2 \@extra@b@citeb \@ifundefined br@#2\@extra@b@citeb \@namedef br@#2 \@nameuse br@#2\@extra@b@citeb \@ifundefined b@#2\@extra@b@citeb @num @parse #2 @tmp #1 NAT@b@open@#2 NAT@b@shut@#2 \@ifnum @merge>\@ne @bibitem@firs...

  40. [48]

    @open @close @open @close and [1] URL: #1 \@ifundefined chapter * \@mkboth \@ifxundefined @sectionbib * \@mkboth * \@mkboth\@gobbletwo \@ifclassloaded amsart * \@ifclassloaded amsbook * \@ifxundefined @heading @heading NAT@ctr thebibliography [1] @ \@biblabel @NAT@ctr \@bibset...

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.