Learning to Count in the Crowd from Limited Labeled Data

Sindagi, Vishwanath A.; Yasarla, Rajeev; Sam, Deepak Babu; Babu, R. Venkatesh; Patel, Vishal M.

doi:10.1007/978-3-030-58621-8_13

Cited by 46 publications

(15 citation statements)

References 51 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Another exception is the method of [78]. It uses pseudo labels generated by a network trained on synthetic images as though they were ground-truth labels.…”

Section: Related Workmentioning

confidence: 99%

“…Previous works in crowd density estimation use the mean absolute error (M AE) and the root mean squared error (RM SE) as evaluation metrics [86,78]. They are defined as…”

Section: Evaluation Metricsmentioning

confidence: 99%

“…It is a real image dataset that contains only 50 images with a people count ranging from 94 to 4,543, which makes it challenging for a deep-learning approach. For a fair comparison, we use the same 5-fold crossvalidation protocol as in [86,78]: We partition the images into 5 10-image groups. In turn, we then pick four groups for training and the remaining one for testing.…”

Section: Benchmark Datasetsmentioning

confidence: 99%

“…For a fair comparison with previous work [86,78], we use SFCN [86] as the crowd density regressor and Adam [30] for parameter update with a learning rate of 1e − 6. After a grid search on one single dataset as discussed below, we set λ 1 in Eq.…”

Section: Implementation Detailsmentioning

confidence: 99%

“…We consistently and clearly outperform all other methods on all the datasets. And, since we use the same SFCN network archi- tecture as the methods of [86,78], the performance boost is directly attributable to our approach of domain adaptation.…”

Section: Comparing Against Recent Techniquesmentioning

confidence: 99%

See 4 more Smart Citations

Leveraging Self-Supervision for Cross-Domain Crowd Counting

Liu¹,

Durasov²,

Fua³

2021

Preprint

View full text Add to dashboard Cite

State-of-the-art methods for counting people in crowded scenes rely on deep networks to estimate crowd density. While effective, these data-driven approaches rely on large amount of data annotation to achieve good performance, which stops these models from being deployed in emergencies during which data annotation is either too costly or cannot be obtained fast enough.One popular solution is to use synthetic data for training. Unfortunately, due to domain shift, the resulting models generalize poorly on real imagery. We remedy this shortcoming by training with both synthetic images, along with their associated labels, and unlabeled real images. To this end, we force our network to learn perspective-aware features by training it to recognize upside-down real images from regular ones and incorporate into it the ability to predict its own uncertainty so that it can generate useful pseudo labels for fine-tuning purposes. This yields an algorithm that consistently outperforms state-of-the-art cross-domain crowd counting ones without any extra computation at inference time.

show abstract

“…Another exception is the method of [78]. It uses pseudo labels generated by a network trained on synthetic images as though they were ground-truth labels.…”

Section: Related Workmentioning

confidence: 99%

“…Previous works in crowd density estimation use the mean absolute error (M AE) and the root mean squared error (RM SE) as evaluation metrics [86,78]. They are defined as…”

Section: Evaluation Metricsmentioning

confidence: 99%

Section: Benchmark Datasetsmentioning

confidence: 99%

Section: Implementation Detailsmentioning

confidence: 99%

Section: Comparing Against Recent Techniquesmentioning

confidence: 99%

See 3 more Smart Citations

Leveraging Self-Supervision for Cross-Domain Crowd Counting

Liu¹,

Durasov²,

Fua³

2021

Preprint

View full text Add to dashboard Cite

show abstract

Region feature smoothness assumption for weakly semi‐supervised crowd counting

Zhang

Piao

et al. 2023

Computer Animation & Virtual

View full text Add to dashboard Cite

Crowd counting is a hot issue in visual data processing. It also plays an important role in the field of video surveillance, social security, and traffic control. However, most of the existing crowd counting methods always adopt a mount of training data or point‐level annotation to learn the mapping relationships between images and density maps, which would cost much human labor. In this paper, we propose a new weakly semi‐supervised crowd counting method which uses less count‐level data for data training. In particular, we extend the classical smoothness assumption and design a many‐to‐many Region Feature Smoothness Assumption to deal with the uneven density distribution problem within crowd region. Further, we adopt hypergraph representation to explore the complex high‐order relationship for different crowd regions. Besides, we design a multi‐scale dynamic hypergraph convolutional module and hyperedge contrastive loss. Extensive experiments have been conducted on five public datasets. The experimental results show that the proposed method outperforms the state‐of‐the‐art ones.

show abstract

Prior-Based Domain Adaptive Object Detection for Hazy and Rainy Conditions

Sindagi

Oza

Yasarla

et al. 2020

Computer Vision – ECCV 2020

Self Cite

View full text Add to dashboard Cite

Adverse weather conditions such as rain and haze corrupt the quality of captured images, which cause detection networks trained on clean images to perform poorly on these images. To address this issue, we propose an unsupervised prior-based domain adversarial object detection framework for adapting the detectors to different weather conditions. We make the observations that corruptions due to different weather conditions (i) follow the principles of physics and hence, can be mathematically modeled, and (ii) often cause degradations in the feature space leading to deterioration in the detection performance. Motivated by these, we propose to use weather-specific prior knowledge obtained using the principles of image formation to define a novel prior-adversarial loss. The prior-adversarial loss used to train the adaptation process aims to produce weather-invariant features by reducing the weather-specific information in the features, thereby mitigating the effects of weather on the detection performance. Additionally, we introduce a set of residual feature recovery blocks in the object detection pipeline to de-distort the feature space, resulting in further improvements. The proposed framework outperforms all existing methods by a large margin when evaluated on different datasets such as Foggy-Cityscapes, Rainy-Cityscapes, RTTS and UFDD.

show abstract

Learning to Count in the Crowd from Limited Labeled Data

Cited by 46 publications

References 51 publications

Leveraging Self-Supervision for Cross-Domain Crowd Counting

Leveraging Self-Supervision for Cross-Domain Crowd Counting

Region feature smoothness assumption for weakly semi‐supervised crowd counting

Prior-Based Domain Adaptive Object Detection for Hazy and Rainy Conditions

Contact Info

Product

Resources

About