Unsupervised Crowd Counting

Elassal, Nada; Elder, James H.

doi:10.1007/978-3-319-54193-8_21

Search citation statements

Order By: Relevance

Paper Sections

Select...

A Crowd Counting1

Counting For Scarce Labeled Data1

Citation Types

Supporting

Mentioning

Contrasting

Year Published

2017

2023

Publication Types

Select...

Other5

Article2

Relationship

Self Cite0

Independent7

Authors

Journals

Cited by 7 publications

(2 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Besides the supervised methods, several approaches focus on relieving the labeling burdensome. They can be broadly categorized into semisupervised methods [5], [40], [55], [68], [91], [93], weaklysupervised methods [83], self-supervised methods [38], [39] and unsupervised methods [14], [60].…”

Section: A Crowd Countingmentioning

confidence: 99%

Fine-grained Domain Adaptive Crowd Counting via Point-derived Segmentation

Liu¹,

Xu²,

Ren³

et al. 2021

Preprint

View full text Add to dashboard Cite

Section: A Crowd Countingmentioning

confidence: 99%

Fine-grained Domain Adaptive Crowd Counting via Point-derived Segmentation

Liu¹,

Xu²,

Ren³

et al. 2021

Preprint

View full text Add to dashboard Cite

“…Due to the high cost of manually labeling data, some works [8,25,33,42] attempt to tackle this problems. Elassal and Elder [8] are the first to propose an unsupervised crowd counting via automatically learning how many groups of people. However, it cannot perform well in the congested crowd scenes.…”

Section: Counting For Scarce Labeled Datamentioning

confidence: 99%

Feature-aware Adaptation and Density Alignment for Crowd Counting in Video Surveillance

Yuan

Wang

2019

Preprint

View full text Add to dashboard Cite

With the development of deep neural networks, the performance of crowd counting and pixel-wise density estimation are continually being refreshed. Despite this, there are still two challenging problems in this field: 1) current supervised learning needs a large amount of training data, but collecting and annotating them is difficult; 2) existing methods can not generalize well to the unseen domain. A recently released synthetic crowd dataset alleviates these two problems. However, the domain gap between the real-world data and synthetic images decreases the models' performance. To reduce the gap, in this paper, we propose a domain-adaptation-style crowd counting method, which can effectively adapt the model from synthetic data to the specific real-world scenes. It consists of Multi-level Feature-aware Adaptation (MFA) and Structured Density map Alignment (SDA). To be specific, MFA boosts the model to extract domain-invariant features from multiple layers. SDA guarantees the network outputs fine density maps with a reasonable distribution on the real domain. Finally, we evaluate the proposed method on four mainstream surveillance crowd datasets, Shanghai Tech Part B, WorldExpo'10, Mall and UCSD. Extensive experiments evidence that our approach outperforms the state-of-the-art methods for the same cross-domain counting problem.

show abstract

Estimating Camera Tilt from Motion without Tracking

Elassal

Elder

2017

2017 14th Conference on Computer and Robot Vision (CRV)

View full text Add to dashboard Cite

Unsupervised Crowd Counting

Cited by 7 publications

References 38 publications

Fine-grained Domain Adaptive Crowd Counting via Point-derived Segmentation

Fine-grained Domain Adaptive Crowd Counting via Point-derived Segmentation

Feature-aware Adaptation and Density Alignment for Crowd Counting in Video Surveillance

Estimating Camera Tilt from Motion without Tracking

Contact Info

Product

Resources

About