ProgRoCC: A Progressive Approach to Rough Crowd Counting

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Jiang, Shengqin, Li, Linfei, Zhang, Haokui, Liu, Qingshan, Beheshti, Amin, Yang, Jian, Hengel, Anton van den, Sheng, Quan Z., Qi, Yuankai
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866913799273971712
author Jiang, Shengqin
Li, Linfei
Zhang, Haokui
Liu, Qingshan
Beheshti, Amin
Yang, Jian
Hengel, Anton van den
Sheng, Quan Z.
Qi, Yuankai
author_facet Jiang, Shengqin
Li, Linfei
Zhang, Haokui
Liu, Qingshan
Beheshti, Amin
Yang, Jian
Hengel, Anton van den
Sheng, Quan Z.
Qi, Yuankai
contents As the number of individuals in a crowd grows, enumeration-based techniques become increasingly infeasible and their estimates increasingly unreliable. We propose instead an estimation-based version of the problem: we label Rough Crowd Counting that delivers better accuracy on the basis of training data that is easier to acquire. Rough crowd counting requires only rough annotations of the number of targets in an image, instead of the more traditional, and far more expensive, per-target annotations. We propose an approach to the rough crowd counting problem based on CLIP, termed ProgRoCC. Specifically, we introduce a progressive estimation learning strategy that determines the object count through a coarse-to-fine approach. This approach delivers answers quickly, outperforms the state-of-the-art in semi- and weakly-supervised crowd counting. In addition, we design a vision-language matching adapter that optimizes key-value pairs by mining effective matches of two modalities to refine the visual features, thereby improving the final performance. Extensive experimental results on three widely adopted crowd counting datasets demonstrate the effectiveness of our method.
format Preprint
id arxiv_https___arxiv_org_abs_2504_13405
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle ProgRoCC: A Progressive Approach to Rough Crowd Counting
Jiang, Shengqin
Li, Linfei
Zhang, Haokui
Liu, Qingshan
Beheshti, Amin
Yang, Jian
Hengel, Anton van den
Sheng, Quan Z.
Qi, Yuankai
Computer Vision and Pattern Recognition
As the number of individuals in a crowd grows, enumeration-based techniques become increasingly infeasible and their estimates increasingly unreliable. We propose instead an estimation-based version of the problem: we label Rough Crowd Counting that delivers better accuracy on the basis of training data that is easier to acquire. Rough crowd counting requires only rough annotations of the number of targets in an image, instead of the more traditional, and far more expensive, per-target annotations. We propose an approach to the rough crowd counting problem based on CLIP, termed ProgRoCC. Specifically, we introduce a progressive estimation learning strategy that determines the object count through a coarse-to-fine approach. This approach delivers answers quickly, outperforms the state-of-the-art in semi- and weakly-supervised crowd counting. In addition, we design a vision-language matching adapter that optimizes key-value pairs by mining effective matches of two modalities to refine the visual features, thereby improving the final performance. Extensive experimental results on three widely adopted crowd counting datasets demonstrate the effectiveness of our method.
title ProgRoCC: A Progressive Approach to Rough Crowd Counting
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2504.13405