KlingTeam/CameraClone-Dataset

Camera Clone Dataset是一个用于相机克隆学习的大规模合成数据集，包含多样化的场景、主题和相机运动。该数据集由三个视频组成：相机运动参考视频、内容参考视频和目标视频，目标视频使用与相机运动参考视频相同的运动方式重新捕捉内容参考视频的场景。数据集使用Unreal Engine 5渲染，并包含了40个3D场景、66个角色模型和93种动画，相机轨迹满足同时多视角捕捉和配对轨迹的要求。

The Camera Clone Dataset is a large-scale synthetic dataset designed for camera clone learning, encompassing diverse scenes, subjects, and camera movements. It consists of three videos: a camera motion reference video, a content reference video, and a target video, which recaptures the scene in the content reference video with the same camera movement as the camera motion reference video. The dataset is rendered using Unreal Engine 5 and includes 40 3D scenes, 66 character models, and 93 different animations. The camera trajectories meet the requirements of simultaneous multi-view capture and paired trajectories.