Iterative Crowd Counting

Ranjan, Viresh; Le, Hieu Van; Hoai, Minh

doi:10.1007/978-3-030-01234-2_17

Cited by 250 publications

(198 citation statements)

References 18 publications

Supporting

Mentioning

196

Contrasting

Order By: Relevance

“…On the other hand, it indicates that existing multi-column networks tend to overfit the data and can not learn the essence of the ground truth. Inspired by previous works [30,44,62], we reveal that the problem of existing multi-column networks lies in the difficulty of learning features with different scales. Generally speaking, there are two main problems: 1) There is no supervision to guide multiple columns to learn features at different scales.…”

Section: Introductionmentioning

confidence: 82%

“…These state-of-the-art methods aim to improve the scale invariance of feature learning. Inspired by recent studies [30,44,62], we reveal that existing multi-column networks cannot effectively learn different scale features as Sec. 1.…”

Section: Cnn-based Methodsmentioning

confidence: 97%

“…Recently, Liu et al [32] propose a multi-column network to simultaneously estimate crowd density by detection and regression models. Ranjan et al [44] employ a two-column structure to iterative train their model with different resolution images.…”

Section: Cnn-based Methodsmentioning

confidence: 99%

“…To solve this problem, we propose a novel Multi-column Mutual Learning (McML) strategy, which can be applied to all existing CNN-based multi-column networks and is end-to-end trainable. It is noted that the previous work ic-CNN [44] also proposes an iterative learning strategy to improve the learning of multi-column networks. Different from our McML, since ic-CNN is designed for a specific neural architecture, it can not be generalized to all multi-column networks.…”

Section: Cnn-based Methodsmentioning

confidence: 99%

See 3 more Smart Citations

Improving the Learning of Multi-column Convolutional Neural Network for Crowd Counting

Cheng

Dai

et al. 2019

Proceedings of the 27th ACM International Conference on Multimedia

View full text Add to dashboard Cite

Tremendous variation in the scale of people/head size is a critical problem for crowd counting. To improve the scale invariance of feature representation, recent works extensively employ Convolutional Neural Networks with multi-column structures to handle different scales and resolutions. However, due to the substantial redundant parameters in columns, existing multi-column networks invariably exhibit almost the same scale features in different columns, which severely affects counting accuracy and leads to overfitting. In this paper, we attack this problem by proposing a novel Multicolumn Mutual Learning (McML) strategy. It has two main innovations: 1) A statistical network is incorporated into the multi-column framework to estimate the mutual information between columns, which can approximately indicate the scale correlation between features from different columns. By minimizing the mutual information, each column is guided to learn features with different image scales. 2) We devise a mutual learning scheme that can alternately optimize each column while keeping the other columns fixed on each mini-batch training data. With such asynchronous parameter update process, each column is inclined to learn different feature representation from others, which can efficiently reduce the parameter redundancy and improve generalization ability. More remarkably, McML can be applied to all existing multi-column networks and is end-to-end trainable. Extensive experiments on four challenging benchmarks show that McML can significantly improve the original multi-column networks and outperform the other state-of-the-art approaches.

show abstract

Section: Introductionmentioning

confidence: 82%

Section: Cnn-based Methodsmentioning

confidence: 97%

Section: Cnn-based Methodsmentioning

confidence: 99%

Section: Cnn-based Methodsmentioning

confidence: 99%

See 2 more Smart Citations

Improving the Learning of Multi-column Convolutional Neural Network for Crowd Counting

Cheng

Dai

et al. 2019

Proceedings of the 27th ACM International Conference on Multimedia

View full text Add to dashboard Cite

show abstract

“…In contrast to these methods that emphasize on specifically addressing large-scale variations in head sizes, the most recent methods ( [2] , [39], [41], [24], [33]) have focused on other properties of the problem. For instance, Babu et al [2] proposed a mechanism to incrementally increase the network capacity conditioned on the dataset.…”

Section: Related Workmentioning

confidence: 99%

Pushing the Frontiers of Unconstrained Crowd Counting: New Dataset and Benchmark Method

Sindagi

Yasarla

Patel

2019

2019 IEEE/CVF International Conference on Computer Vision (ICCV)

View full text Add to dashboard Cite

In this work, we propose a novel crowd counting network that progressively generates crowd density maps via residual error estimation. The proposed method uses VGG16 as the backbone network and employs density map generated by the final layer as a coarse prediction to refine and generate finer density maps in a progressive fashion using residual learning. Additionally, the residual learning is guided by an uncertainty-based confidence weighting mechanism that permits the flow of only high-confidence residuals in the refinement path. The proposed Confidence Guided Deep Residual Counting Network (CG-DRCN) is evaluated on recent complex datasets, and it achieves significant improvements in errors.Furthermore, we introduce a new large scale unconstrained crowd counting dataset (JHU-CROWD) that is ∼2.8 × larger than the most recent crowd counting datasets in terms of the number of images. It contains 4,250 images with 1.11 million annotations. In comparison to existing datasets, the proposed dataset is collected under a variety of diverse scenarios and environmental conditions. Specifically, the dataset includes several images with weatherbased degradations and illumination variations in addition to many distractor images, making it a very challenging dataset. Additionally, the dataset consists of rich annotations at both image-level and head-level. Several recent methods are evaluated and compared on this dataset.

show abstract

Class-Agnostic Counting

Xie

Zisserman

2019

Computer Vision – ACCV 2018

View full text Add to dashboard Cite

Nearly all existing counting methods are designed for a specific object class. Our work, however, aims to create a counting model able to count any class of object. To achieve this goal, we formulate counting as a matching problem, enabling us to exploit the image selfsimilarity property that naturally exists in object counting problems. We make the following three contributions: first, a Generic Matching Network (GMN) architecture that can potentially count any object in a class-agnostic manner; second, by reformulating the counting problem as one of matching objects, we can take advantage of the abundance of video data labeled for tracking, which contains natural repetitions suitable for training a counting model. Such data enables us to train the GMN. Third, to customize the GMN to different user requirements, an adapter module is used to specialize the model with minimal effort, i.e. using a few labeled examples, and adapting only a small fraction of the trained parameters. This is a form of few-shot learning, which is practical for domains where labels are limited due to requiring expert knowledge (e.g. microbiology). We demonstrate the flexibility of our method on a diverse set of existing counting benchmarks: specifically cells, cars, and human crowds. The model achieves competitive performance on cell and crowd counting datasets, and surpasses the state-of-the-art on the car dataset using only three training images. When training on the entire dataset, the proposed method outperforms all previous methods by a large margin.

show abstract

Iterative Crowd Counting

Cited by 250 publications

References 18 publications

Improving the Learning of Multi-column Convolutional Neural Network for Crowd Counting

Improving the Learning of Multi-column Convolutional Neural Network for Crowd Counting

Pushing the Frontiers of Unconstrained Crowd Counting: New Dataset and Benchmark Method

Class-Agnostic Counting

Contact Info

Product

Resources

About