Learning to Count in the Crowd from Limited Labeled Data

@inproceedings{Sindagi2020LearningTC,
  title={Learning to Count in the Crowd from Limited Labeled Data},
  author={Vishwanath A. Sindagi and Rajeev Yasarla and Deepak Sam Babu and R. Venkatesh Babu and Vishal M. Patel},
  booktitle={ECCV},
  year={2020}
}
Recent crowd counting approaches have achieved excellent performance. However, they are essentially based on fully supervised paradigm and require large number of annotated samples. Obtaining annotations is an expensive and labour-intensive process. In this work, we focus on reducing the annotation efforts by learning to count in the crowd from limited number of labeled samples while leveraging a large pool of unlabeled data. Specifically, we propose a Gaussian Process-based iterative learning… 
Leveraging Self-Supervision for Cross-Domain Crowd Counting
TLDR
This work forces a 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.
Self-supervised Domain Adaptation in Crowd Counting
TLDR
This work introduces a new approach to utilize existing datasets with ground truth to produce more robust predictions on unlabeled datasets, named domain adaptation, in crowd counting, where the network is trained with labeled data and samples without labels are also added to the training process.
S4-Crowd: Semi-Supervised Learning with Self-Supervised Regularisation for Crowd Counting
TLDR
A semi-supervised learning framework S4-C Crowd is proposed, which can leverage both unlabeled/labeled data for robust crowd modelling and a crowd-driven recurrent unit Gated-Crowd-RecurrentUnit (GCRU), which can preserve discriminant crowd information by extracting second-order statistics, yielding pseudo labels with improved quality.
S2FPR: Crowd Counting via Self-Supervised Coarse to Fine Feature Pyramid Ranking
TLDR
This work proposes SFPR, a semisupervised crowd counting method which can extract structural information and learn partial orders of coarse-to-fine pyramid features in the latent space for better crowd counting with massive unlabeled images.
Bi-level Alignment for Cross-Domain Crowd Counting
TLDR
This work designs a bi-level alignment framework (BLA) consisting of task-driven data alignment and feature alignment, and introduces Au-toML to search for an optimal transform on source, which well serves for the downstream task.
CrowdMLP: Weakly-Supervised Crowd Counting via Multi-Granularity MLP
TLDR
A novel and efficient counter that probes into modelling global dependencies of embeddings and regressing total counts by devising a multi-granularity MLP regressor and an effective proxy task, namely Split-Counting, is proposed to evade the barrier of limited samples and the shortage of spatial hints in a self-supervised manner.
Crowd Counting With Partial Annotations in an Image
TLDR
This paper studies a novel crowd counting setting, i.e. only using partial annotations in each image as training data, which achieves better performance than the recent methods and baselines under semi-supervised or active learning settings on all datasets.
Spatial Uncertainty-Aware Semi-Supervised Crowd Counting
TLDR
This paper proposes a spatial uncertainty-aware semi-supervised approach via regularized surrogate task (binary segmentation) for crowd counting problems and estimates the spatial uncertainty maps from the teacher model’s surrogate task to guide the feature learning of the main task (density regression) and the surrogate task of the student model at the same time.
Reducing Spatial Labeling Redundancy for Semi-supervised Crowd Counting
TLDR
This work proposes to break the labeling chain of previous methods and make the first attempt to reduce spatial labeling redundancy for semi-supervised crowd counting, and exploits the similarities among individuals in each crowd image to directly supervise the unlabeled regions via feature propagation.
Fine-grained Domain Adaptive Crowd Counting via Point-derived Segmentation
TLDR
This work proposes a novel and effective schema to learn crowd segmentation from point-level crowd counting annotations in the context of Multiple Instance Learning and proposes a crowd-aware fine-grained domain adaptation framework for crowd counting.
...
...

References

SHOWING 1-10 OF 64 REFERENCES
Learning From Synthetic Data for Crowd Counting in the Wild
TLDR
A data collector and labeler is developed which can generate the synthetic crowd scenes and simultaneously annotate them without any manpower, and a crowd counting method via domain adaptation is proposed, which can free humans from heavy data annotations.
From Semi-supervised to Transfer Counting of Crowds
TLDR
This study proposes a unified active and semi-supervised regression framework with ability to perform transfer learning, by exploiting the underlying geometric structure of crowd patterns via manifold analysis.
Almost Unsupervised Learning for Dense Crowd Counting
TLDR
Grid Winner-Take-All (GWTA) autoencoder is developed to learn several layers of useful filters from unlabeled crowd images to achieve superior results compared to other unsupervised methods and stays reasonably close to the accuracy of supervised baseline.
Leveraging Unlabeled Data for Crowd Counting by Learning to Rank
TLDR
This work proposes a novel crowd counting approach that leverages abundantly available unlabeled crowd imagery in a learning-to-rank framework and demonstrates how to efficiently learn from these unlabeling datasets by incorporating learning- to-rank in a multi-task network which simultaneously ranks images and estimates crowd density maps.
Realistic Evaluation of Deep Semi-Supervised Learning Algorithms
TLDR
This work creates a unified reimplemention and evaluation platform of various widely-used SSL techniques and finds that the performance of simple baselines which do not use unlabeled data is often underreported, that SSL methods differ in sensitivity to the amount of labeled and unlabeling data, and that performance can degrade substantially when the unlabelED dataset contains out-of-class examples.
Pushing the Frontiers of Unconstrained Crowd Counting: New Dataset and Benchmark Method
TLDR
A novel crowd counting network that progressively generates crowd density maps via residual error estimation using 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.
Residual Regression With Semantic Prior for Crowd Counting
TLDR
A residual regression framework is proposed for crowd counting utilizing the correlation information among samples, and it is discovered that more intrinsic characteristics can be learned by the network which thus generalizes better to unseen scenarios.
Syn2Real Transfer Learning for Image Deraining Using Gaussian Processes
TLDR
This work proposes a Gaussian Process-based semi-supervised learning framework which enables the network in learning to derain using synthetic dataset while generalizing better using unlabeled real-world images.
Context-Aware Crowd Counting
TLDR
This paper introduces an end-to-end trainable deep architecture that combines features obtained using multiple receptive field sizes and learns the importance of each such feature at each image location, which yields an algorithm that outperforms state-of-the-art crowd counting methods, especially when perspective effects are strong.
Cross-scene crowd counting via deep convolutional neural networks
TLDR
A deep convolutional neural network is proposed for crowd counting, and it is trained alternatively with two related learning objectives, crowd density and crowd count, to obtain better local optimum for both objectives.
...
...