宠物猫狗数据集

数据包含猫狗图片共计40G，包含92457张猫狗照片。猫、狗数据采集场景为室内与室外结合的真实场景，其中室内场景占比超过80%。该数据涵盖多种室内场景、多种猫狗类型、多种拍摄角度、多种距离和多种光照条件。在标注方面，对猫、狗位置进行矩形框标注，并添加标签分类。此猫狗数据集，可以帮开发人员进行目标检测、分类和识别等任务，用户可以训练和优化模型、优化猫狗的身份认证系统。产品名称宠物猫狗数据集数据量92457总框数90000+文件格式jpg/xml，png/xml，jpg/txt采集环境室内、室外光照条件自然光采集设备手机拍摄距离近-远拍摄角度多角度准确率检测框贴合度不低于20个像素； 检测框标签准确率不低于95%。

This dataset contains 40GB of cat and dog images, with a total of 92,457 photos. The data was collected in combined indoor and outdoor real-world scenarios, where indoor scenarios account for over 80% of the total. It covers diverse indoor scenes, various cat and dog breeds, multiple shooting angles, shooting distances and lighting conditions. In terms of annotation, rectangular bounding boxes are used to mark the positions of cats and dogs, with corresponding category labels added. This cat and dog dataset can assist developers in conducting tasks such as object detection, classification and recognition, allowing users to train and optimize models as well as optimize pet cat and dog identity authentication systems. The detailed specifications of the dataset are as follows: Product name: Pet Cat and Dog Dataset; Total sample count: 92,457; Total number of bounding boxes: over 90,000; Supported file formats: jpg/xml, png/xml, jpg/txt; Collection environment: indoor and outdoor with natural lighting; Acquisition equipment: mobile phones; Shooting distance range: from near to far; Shooting angles: multi-angle; Annotation accuracy requirements: The bounding box fitting degree is no less than 20 pixels, and the accuracy rate of bounding box labels is no less than 95%.

搜集汇总

背景与挑战

背景概述

该宠物猫狗数据集包含92,457张室内外场景的猫狗图片，其中80%为室内场景，每张图片均带有矩形框标注和分类标签，适用于目标检测与识别模型训练。数据涵盖多种品种、角度和光照条件，标注准确率达95%以上。