PascalVOC Dataset
收藏arXiv2025-09-30 收录
下载链接：
https://byungki-k.github.io/JointDiT/
下载链接
链接失效反馈官方服务：
资源简介：
该数据集包含了来自PascalVOC数据集的16,000张图片，这些图片与深度图和文本提示一一对应。深度图来源于Depth-Anything-V2，而文本提示则是通过BLIP2提取的。该数据集的规模为16,000张图像，旨在支持深度条件下的图像生成任务。
This dataset comprises 16,000 images sourced from the PascalVOC dataset, where each image is paired with a corresponding depth map and text prompt. The depth maps are generated via Depth-Anything-V2, and the text prompts are extracted using BLIP2. With a total of 16,000 images, this dataset is designed to support depth-conditioned image generation tasks.
