• Corpus ID: 202716138

One-Shot Scene-Specific Crowd Counting

@inproceedings{Hossain2019OneShotSC,
  title={One-Shot Scene-Specific Crowd Counting},
  author={Mohammad Asiful Hossain and Mahesh Kumar Krishna Reddy and M. Hosseinzadeh and Omit Chanda and Yang Wang},
  booktitle={BMVC},
  year={2019}
}
We consider the problem of crowd counting in static images. Given an image, the goal is to estimate a density map of this image, where each value in the density map indicates the density level of the corresponding location in the image. In particular, we consider a novel problem setting which we call the one-shot scene-specific crowd counting. During training, we assume that we have labeled images collected from different scenes. Each scene corresponds to a camera at a fixed location and angle… 

Figures and Tables from this paper

Few-Shot Scene Adaptive Crowd Counting Using Meta-Learning
TLDR
The proposed approach learns the model parameters in a way that facilitates the fast adaptation to the target scene with a few gradient updates to the learned parameters and outperforms other alternatives in few-shot scene adaptive crowd counting.
Domain Adaptation in Crowd Counting
TLDR
This paper proposes several domain adaptation techniques for crowd counting that are closer to real-world deployment of crowd counting systems and demonstrates the superior performance of the proposed approach on several benchmark datasets.
AdaCrowd: Unlabeled Scene Adaptation for Crowd Counting
TLDR
This paper proposes a new problem called unlabeled scene-adaptive crowd counting, and introduces a novel AdaCrowd framework to solve this problem, which consists of a crowd counting network and a guiding network that allows the model to adapt to different target scenes.
Meta-Knowledge and Multi-Task Learning-Based Multi-Scene Adaptive Crowd Counting
TLDR
A multi-scene adaptive crowd counting method based on meta-knowledge and multi-task learning that can be deployed to multiple new scenes without duplicated model training and reduces the MAE by 10.29% and 13.48%, respectively.
Active Crowd Counting with Limited Supervision
TLDR
An active learning framework which enables accurate crowd counting with limited supervision is presented: given a small labeling budget, instead of randomly selecting images to annotate, an active labeling strategy is introduced to annotated the most informative images in the dataset and learn the counting model upon them.
A Flow Base Bi-path Network for Cross-scene Video Crowd Understanding in Aerial View
TLDR
A double-stream crowd counting model is proposed, which extracts optical flow and frame difference information as an additional branch and synthetic data generated by game Grand Theft Auto V is introduced to tackle the crowd density estimation problem under extreme dark environments.
CNN-based Density Estimation and Crowd Counting: A Survey
TLDR
Over 220 works are surveyed to comprehensively and systematically study the crowd counting models, mainly CNN-based density map estimation methods to make reasonable inference and prediction for the future development of crowd counting and to provide feasible solutions for the problem of object counting in other fields.
Towards Learning Multi-domain Crowd Counting
TLDR
A domainguided channel attention network (DCANet) towards learning multi-domain crowd counting, which consists of feature extraction module, channel attention-guided multidilation (CAMD) module and density map prediction module, and which performs favorably for handling multi- domain datasets using one single model.
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.
Towards Unsupervised Crowd Counting via Regression-Detection Bi-knowledge Transfer
TLDR
This paper introduces an iterative self-supervised learning scheme with regression-detection bi-knowledge transfer in the target and formulate the mutual transformations between the outputs of regression- and detection-based models as two scene-agnostic transformers which enable knowledge distillation between the two models.
...
...

References

SHOWING 1-10 OF 20 REFERENCES
Single-Image Crowd Counting via Multi-Column Convolutional Neural Network
TLDR
With the proposed simple MCNN model, the method outperforms all existing methods and experiments show that the model, once trained on one dataset, can be readily transferred to a new dataset.
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.
End-to-end crowd counting via joint learning local and global count
  • C. Shang, H. Ai, Bo Bai
  • Computer Science
    2016 IEEE International Conference on Image Processing (ICIP)
  • 2016
TLDR
An end-to-end convolutional neural network architecture that takes a whole image as its input and directly outputs the counting result, taking advantages of contextual information when predicting both local and global count is proposed.
DecideNet: Counting Varying Density Crowds Through Attention Guided Detection and Density Estimation
TLDR
A novel end-to-end crowd counting framework, named DecideNet (DEteCtIon and Density Estimation Network) is proposed, which can adaptively decide the appropriate counting mode for different locations on the image based on its real density conditions.
Switching Convolutional Neural Network for Crowd Counting
TLDR
A novel crowd counting model that maps a given crowd scene to its density and switch convolutional neural network that leverages variation of crowd density within an image to improve the accuracy and localization of the predicted crowd count is proposed.
Feature Mining for Localised Crowd Counting
TLDR
This paper presents a single regression model based approach that is able to estimate people count in spatially localised regions and is more scalable without the need for training a large number of regressors proportional to the number of local regions.
CNN-Based cascaded multi-task learning of high-level prior and density estimation for crowd counting
TLDR
A novel end-to-end cascaded network of CNNs to jointly learn crowd count classification and density map estimation achieves lower count error and better quality density maps as compared to the recent state-of-the-art methods.
Extremely Overlapping Vehicle Counting
TLDR
The challenging problem that this paper is to precisely estimate the number of vehicles in an image of a traffic congestion situation is explored and TRANCOS, a novel database for extremely overlapping vehicle counting, is introduced.
One-shot learning of object categories
TLDR
It is found that on a database of more than 100 categories, the Bayesian approach produces informative models when the number of training examples is too small for other methods to operate successfully.
One-Shot Video Object Segmentation
TLDR
One-Shot Video Object Segmentation (OSVOS), based on a fully-convolutional neural network architecture that is able to successively transfer generic semantic information, learned on ImageNet, to the task of foreground segmentation, and finally to learning the appearance of a single annotated object of the test sequence (hence one-shot).
...
...