Amar-S/LLNL_DSC_2025

LLNL 2025数据科学挑战数据集是一个与图像和视频处理相关的数据集，包含深度估计、图像特征提取、图像转视频、遮罩生成、对象检测、零样本对象检测等任务。该数据集支持多种摄像头、视频对象分割、可提示分割、隐式内容分割、隐式对象表示和对象检索等标签。数据集适用于1M<n<10M的大小类别，并使用英文作为其主要语言。

The LLNL 2025 Data Science Challenge dataset is related to image and video processing, including tasks such as depth estimation, image feature extraction, image-to-video conversion, mask generation, object detection, and zero-shot object detection. The dataset supports tags like multicamera, video object segmentation, promptable segmentation, amodal segmentation, amodal content, amodal object representations, and object retrieval. It falls under the size category of 1M<n<10M and uses English as its primary language.