Scene-Adaptive Attention Network for Crowd Counting

Abstract

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.

RGB-T Multi-Modal Crowd Counting Based on Transformer

“…BCCTrans [28] introduces a global context learnable token to guide the counting. SAANet [40] designs a deformer backbone to extract the features, aggregates multi-level features by a deformable transformer encoder, and introduces a count query in a transformer decoder to re-calibrates the multi-level feature maps. DCSwinTrans [10] enhances the large-range contextual information by a dilated Swin Transformer backbone, and equips with a feature pyramid networks decoder to achieve crowd instant localization.…”

“…It is widely studied by the academia and industrial communities since the number of persons is an important indicator of incident monitoring[31], traffic control [19], and infectious disease prevention [32]. The existing crowd counting methods have achieved tremendous improvement due to the introduce of convolutional neural networks [7,8] and transformer [28,40].However, when light is insufficient, the performance of crowd counting is unsatisfying, as shown in the first line of Fig. 1.…”

MAFNet: A Multi-Attention Fusion Network for RGB-T Crowd Counting

“…To alleviate the problem of difficult collection and annotation of crowd counting datasets, some works [15], [16], [18] explore the domain-adaptive crowd counting from the synthetic datasets to the real-world. In addition, with the Vision Transformer (ViT) [3] first applying the transformer structure for vision tasks, many transformer-based [25] crowd counting methods [43]- [45] have been proposed with outstanding performance.…”

Crowd Localization from Gaussian Mixture Scoped Knowledge and Scoped Teacher

“…C ROWD analysis is a popular application of computer vision and has achieved superb success, especially in crowd counting [10], [46], [59]. Crowd counting is a fundamental task, which estimates the sum counts of instances in crowd scenes.…”