Revisiting Feature Prediction for Learning Visual Representations from Video

arXiv:2404.08471v1 Announce Type: cross
Abstract: This paper explores feature prediction as a stand-alone objective for unsupervised learning from video and introduces V-JEPA, a collection of vision models trained solely using a feature prediction objective, without the use of pretrained image encoders, text, negative examples, reconstruction, or other sources of supervision. The models are trained on 2 million videos collected from public datasets and are evaluated on downstream image and video tasks. Our results show that learning by predicting video features …

abstract arxiv collection cs.ai cs.cv cs.lg examples feature image jepa negative paper prediction supervision text type unsupervised unsupervised learning video vision vision models visual v-jepa