ViMoLa: Discrete Motion Tokens for Language-Guided Video Generation
预印本论文，未放权重，无法直接用于视频生成
Language-guided video generation often captures scene content while missing the requested motion. We present ViMoLa (Video Motion Language), which predicts a motion prior from language and injects it into a frozen image-to-video backbone. A spatial vector-quantized autoencoder co
判断口径 agnes-3.0-flash · 入库 #70 · 回首页