MC-JEPA: A Joint-Embedding Predictive Architecture for Self-Supervised Learning of Motion and Content Features

作者：作者信息待提取 · 单位：Meta AI, FAIR, Inria, ´Ecole normale sup´erieure, CNRS, PSL Research University, Courant Institute, New York University, Center for Data Science, New York University · 来源：General Approaches to World Models · 发布日期：2023 · 分类：6. Latent Space World Models