Hugging Face 釋出 vid_ds_scripts：一站式構建影片生成高品質資料集
Original: Build awesome datasets for video generation
With the rise of open-source video generation models such as LTX-Video, HunyuanVideo, and CogVideoX, building high-quality training…
Hugging Face 發表全新開源工具包 vid_ds_scripts，解決影片生成模型（如 LTX-Video、HunyuanVideo）訓練資料準備的痛點。該工具包提供一站式解決方案，涵蓋影片下載、PySceneDetect 場景分割、VLM 自動生成詳細描述，以及資料過濾與格式化。這大幅降低了開發者構建高品質「影片-文字對」資料集的門檻，加速開源影片生成技術的微調與研發。
With the rise of open-source video generation models such as LTX-Video, HunyuanVideo, and CogVideoX, building high-quality training datasets has become the greatest challenge facing developers and researchers. Hugging Face recently released an open-source toolkit and tutorial guide called `vid_ds_scripts`, aimed at simplifying and standardizing the video dataset preparation workflow.
Free shows the 3-line summary; Pro unlocks the full deep summary (~300 words) so you never have to click through.
See Pro plans →Want the original English / full article?
Read on Hugging Face Blog →Summaries are AI-generated; the original article is authoritative.