iWildCam 2021 Competition Dataset

iWildCam 2021数据集是由加州理工学院和Google合作创建，旨在通过自动化的图像分析技术，帮助生态学家更有效地监测全球动物种群。该数据集包含263,528张来自323个不同地点的相机陷阱图像，涉及206种动物。数据集的创建过程包括收集相机陷阱图像、公民科学图像和多光谱图像，并通过专家和众包方式进行标注。该数据集主要用于解决动物种群数量估计和物种识别问题，支持全球范围内的生物多样性研究。

The iWildCam 2021 dataset was collaboratively created by the California Institute of Technology and Google, with the goal of assisting ecologists in more efficiently monitoring global animal populations through automated image analysis techniques. This dataset comprises 263,528 camera trap images sourced from 323 distinct locations, encompassing 206 animal species. The dataset construction workflow includes the collection of camera trap images, citizen science images, and multispectral images, with annotations generated via expert review and crowdsourcing. Primarily designed to tackle challenges in animal population abundance estimation and species identification, this dataset supports global-scale biodiversity research.