Direct Measure Matching for Crowd Counting

Lin, Huei-Yung; Hong, Xiaopeng; Ma, Zhiheng; Wei, Xing; Qiu, Yunfeng; Wang, Yaowei; Gong, Yi

doi:10.24963/ijcai.2021/116

Cited by 32 publications

(12 citation statements)

References 1 publication

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As the result shows, our MAN performs great accuracy on all the four benchmark datasets. MAN improves MAE and MSE values of second best method S3 [14] from 80.6 to 77.3 and from 139.8 to 131.5, respectively. On JHU++, it improves these two values from 59.4 to 53.4 and from 244.0 to 209.9, respectively.…”

Section: Comparison With State-of-the-art Methodsmentioning

confidence: 92%

See 1 more Smart Citation

Boosting Crowd Counting via Multifaceted Attention

Lin¹,

Ma²,

Ji³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

This paper focuses on the challenging crowd counting task. As large-scale variations often exist within crowd images, neither fixed-size convolution kernel of CNN nor fixed-size attention of recent vision transformers can well handle this kind of variations. To address this problem, we propose a Multifaceted Attention Network (MAN) to improve transformer models in local spatial relation encoding. MAN incorporates global attention from vanilla transformer, learnable local attention, and instance attention into a counting model. Firstly, the local Learnable Region Attention (LRA) is proposed to assign attention exclusive for each feature location dynamically. Secondly, we design the Local Attention Regularization to supervise the training of LRA by minimizing the deviation among the attention for different feature locations. Finally, we provide an Instance Attention mechanism to focus on the most important instances dynamically during training. Extensive experiments on four challenging crowd counting datasets namely ShanghaiTech, UCF-QNRF, JHU++, and NWPU have validated the proposed method. Code: https://github.com/LoraLinH/Boosting-Crowd-Counting-via-Multifaceted-Attention.

show abstract

Section: Comparison With State-of-the-art Methodsmentioning

confidence: 92%

“…BL [21] designs the loss function based on Bayesian theory, calculating the deviation of expectation for each crowd. And further works [14,23,38] focus on optimal transport and measure the divergence without depending on the assumption of Gaussian distribution.…”

Section: Crowd Countingmentioning

confidence: 99%

Boosting Crowd Counting via Multifaceted Attention

Lin¹,

Ma²,

Ji³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Appropriate measure matching can help to improve the counting performance. S3 [35] proposes a novel measure matching based on Sinkhorn divergence, avoiding generating the density maps. UOT [36] uses unbalanced optimal transport (UOT) distance to quantify the discrepancy between two measures, outputting sharper density maps.…”

Section: Cnn-based Crowd Countingmentioning

confidence: 99%

TransCrowd: weakly-supervised crowd counting with transformers

Liang

Chen

et al. 2022

Sci. China Inf. Sci.

128

View full text Add to dashboard Cite

The mainstream crowd counting methods usually utilize the convolution neural network (CNN) to regress a density map, requiring point-level annotations. However, annotating each person with a point is an expensive and laborious process. During the testing phase, the point-level annotations are not considered to evaluate the counting accuracy, which means the point-level annotations are redundant. Hence, it is desirable to develop weakly-supervised counting methods that just rely on count-level annotations, a more economical way of labeling. Current weakly-supervised counting methods adopt the CNN to regress a total count of the crowd by an image-to-count paradigm. However, having limited receptive fields for context modeling is an intrinsic limitation of these weakly-supervised CNN-based methods. These methods thus cannot achieve satisfactory performance, with limited applications in the real world. The transformer is a popular sequence-to-sequence prediction model in natural language processing (NLP), which contains a global receptive field. In this paper, we propose TransCrowd, which reformulates the weakly-supervised crowd counting problem from the perspective of sequence-to-count based on transformers. We observe that the proposed TransCrowd can effectively extract the semantic crowd information by using the self-attention mechanism of transformer. To the best of our knowledge, this is the first work to adopt a pure transformer for crowd counting research. Experiments on five benchmark datasets demonstrate that the proposed TransCrowd achieves superior performance compared with all the weakly-supervised CNN-based counting methods and gains highly competitive counting performance compared with some popular fully-supervised counting methods.

show abstract

“…Wang et al [9] formulates crowd counting as a distribution matching problem and constructs loss using optimal transport. Lin et al [45] further improve the loss function based on Sinkhorn distance. More improvements such as incorporating perspective information [41], [46], auxiliary task [34], [47], cross-datasets training [48], [49] and neural architecture search [50] further promote the counting performance.…”

Section: A Crowd Countingmentioning

confidence: 99%

Scene-Adaptive Attention Network for Crowd Counting

Wang¹,

Kang²,

Yang³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

In recent years, significant progress has been made on the research of crowd counting. However, as the challenging scale variations and complex scenes existed in crowds, neither traditional convolution networks nor near recent Transformer architectures with fixed-size attention could handle the task well. To address this problem, this paper proposes a sceneadaptive attention network, termed SAANet. First of all, we design a deformable attention in-built Transformer backbone, which learns adaptive feature representations with deformable sampling locations and dynamic attention weights. Then we propose the multi-level feature fusion and count-attentive feature enhancement modules further, to strengthen feature representation under the global image context. The learned representations could attend to the foreground and are adaptive to different scales of crowds. We conduct extensive experiments on four challenging crowd counting benchmarks, demonstrating that our method achieves state-of-the-art performance. Especially, our method currently ranks No.1 on the public leaderboard of the NWPU-Crowd benchmark. We hope our method could be a strong baseline to support future research in crowd counting. The source code will be released to the community.

show abstract

Direct Measure Matching for Crowd Counting

Cited by 32 publications

References 1 publication

Boosting Crowd Counting via Multifaceted Attention

Boosting Crowd Counting via Multifaceted Attention

TransCrowd: weakly-supervised crowd counting with transformers

Scene-Adaptive Attention Network for Crowd Counting

Contact Info

Product

Resources

About