Scale Aggregation Network for Accurate and Efficient Crowd Counting

By Xinkun Cao, Zhipeng Wang, Yanyun Zhao, Fei Su. European Conference on Computer Vision, 2018

710 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Safety, Risk, Reliability and Quality

View Paper: https://www.semanticscholar.org/paper/bb65479d3f38b73826596e5da9dbf5ee4199d42c

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Xiaopeng Hong
2. Xing Wei
3. Zhongzhan Huang
4. Senwei Liang
5. Yihong Gong
6. Wei He
7. Haizhao Yang
8. Mingfu Liang

In Computer Vision and Pattern Recognition

1. Luc Van Gool
2. Ling Shao
3. Fahad Shahbaz Khan
4. Dacheng Tao
5. Ming-Hsuan Yang
6. Xiaogang Wang
7. Jitendra Malik
8. Andrea Vedaldi

In Safety, Risk, Reliability and Quality

1. Qi Wang
2. Dimitris Samaras
3. Pascal Fua
4. Junyu Gao
5. Wei Lin
6. Weizhe Liu
7. Minh Hoai
8. Fahad Shahbaz Khan

Citations in the corpus

• Context-Aware Crowd Counting (2018, methodology): “The regressors can be based on Random Forests [18], Gaussian Processes [7], or more recently Deep Nets [41, 42, 26, 31, 40, 36, 32, 24, 19, 30, 33, 22, 15, 28, 5], with most state-of-the-art approaches now relying on the latter.”
• Object Counting and Instance Segmentation With Image-Level Supervision (2019, methodology): “Despite being trained using image-level object counts within the subitizing range [1-4], it accurately counts objects beyond the subitizing range (11 persons) under heavy occlusion (marked with blue arrow to show two persons) in the left image and diverse object categories in the right.”
• Deep Structural Contour Detection (2020, neutral): “[51], MCNN [52], Switch CNN [34], ACSCP [37], CSRnet [24], SA-Net [6], MAN [42] and RANet [50].”
• Crowd Counting and Density Estimation by Trellis Encoder-Decoder Networks (2019, methodology): “Most crowd counting methods are based on detection [18, 9, 45], regression [28, 4, 10], and density estimation [17, 53, 2, 11].”
• Efficient Attention Network: Accelerate Attention by Searching Where to Plug (2020, methodology): “To further investigate the generalization of EAN, we conduct experiments on transferring the optimal architecture from image classification to crowd counting task [33, 4, 20, 12], semantic segmentation [9].”
• Distribution Matching for Crowd Counting (2020, methodology): “Almost all prior crowd density map estimation methods [56, 57, 60, 38, 20, 33, 35, 4, 28, 50, 27, 40, 29, 26] have followed this convention.”
• NWPU-Crowd: A Large-Scale Benchmark for Crowd Counting and Localization (2020, neutral): “SANet [33]: Scale Aggregation Network.”
• Learning From Synthetic Data for Crowd Counting in the Wild (2019, methodology): “Traditional models (training from scratch [43, 26, 7] or image classification model [5, 33, 15]) have some layers with random initialization or a regular distribution, which is not a good scheme.”
• Bayesian Loss for Crowd Count Estimation With Point Supervision (2019, background): “5 SANET [4] 67.”