We introduce DIP, a novel unsupervised post-training method designed to enhance dense image representations in large-scale pretrained vision encoders for in-context scene understanding. Unlike prior approaches that rely on complex self-distillation architectures, our method trains the vision encoder using pseudo-tasks that explicitly simulate downstream in-context scenarios, inspired by meta-learning principles. To enable post-training on unlabeled data, we propose an automatic mechanism for generating in-context tasks that combines a pretrained diffusion model and the vision encoder itself. DIP is simple, unsupervised, and computationally efficient, requiring less than 9 hours on a single A100 GPU. By learning dense representations through pseudo in-context tasks, it achieves strong performance across a wide variety of downstream real-world in-context scene understanding tasks. It outperforms both the initial vision encoder and prior methods, offering a practical and effective solution for improving dense representations. Code available here: https://github.com/sirkosophia/DIP
@article{arxiv.2506.18463,
title = {DIP: Unsupervised Dense In-Context Post-training of Visual Representations},
author = {Sophia Sirko-Galouchenko and Spyros Gidaris and Antonin Vobecky and Andrei Bursuc and Nicolas Thome},
journal= {arXiv preprint arXiv:2506.18463},
year = {2025}
}