arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2601.13871cs.CV
收藏

OCCAM: 无训练、无先验、多类别的对象计数

OCCAM: Class-Agnostic, Training-Free, Prior-Free and Multi-Class Object Counting

Computer Science Department, University of Crete(塞萨洛尼基大学计算机科学系);Institute of Computer Science (ICS), Foundation for Research & Technology – Hellas (FORTH)(希腊基础研究与技术研究所计算机科学研究所)

Michail Spanakis, Iason Oikonomidis, Antonis Argyros

AI总结:

OCCAM提出了一种无需训练和先验信息的多类对象计数方法,利用SAM2和FINCH算法在FSC-147和CARPK数据集上实现高性能。

摘要:

类无关的对象计数(CAC)涉及在图像中对任意类别的对象实例进行计数。由于其实际重要性,CAC近年来受到越来越多的关注。大多数现有方法假设每张图像只包含一个对象类别,依赖于对大型深度学习模型的大量训练,并通过引入额外信息(如视觉示例或文本提示)来解决该问题。在本文中,我们提出了OCCAM,这是首个无需训练的CAC方法,无需任何补充信息即可运行。此外,我们的方法解决了多类变体的问题,因为它能够对图像中任意对象类别中的每个类别实例进行计数。我们利用Segment Anything Model 2(SAM2),一个基础模型,以及一种基于阈值的First Integer Neighbor Clustering Hierarchy(FINCH)算法的定制变体,以在广泛使用的基准数据集FSC-147和CARPK上实现具有竞争力的性能。我们提出了一个合成的多类数据集,并提出F1分数作为更合适的评估度量。我们的方法和提出的合成数据集的代码将在https://mikespanak.github.io/OCCAM_counter上公开。

英文摘要:

Class-Agnostic object Counting (CAC) involves counting instances of objects from arbitrary classes within an image. Due to its practical importance, CAC has received increasing attention in recent years. Most existing methods assume a single object class per image, rely on extensive training of large deep learning models and address the problem by incorporating additional information, such as visual exemplars or text prompts. In this paper, we present OCCAM, the first training-free approach to CAC that operates without the need of any supplementary information. Moreover, our approach addresses the multi-class variant of the problem, as it is capable of counting the object instances in each and every class among arbitrary object classes within an image. We leverage Segment Anything Model 2 (SAM2), a foundation model, and a custom threshold-based variant of the First Integer Neighbor Clustering Hierarchy (FINCH) algorithm to achieve competitive performance on widely used benchmark datasets, FSC-147 and CARPK. We propose a synthetic multi-class dataset and F1 score as a more suitable evaluation metric. The code for our method and the proposed synthetic dataset will be made publicly available at https://mikespanak.github.io/OCCAM_counter.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑