Visual Bridge: Universal Visual Perception Representations Generating

TL;DR

This paper introduces a universal visual perception framework based on flow matching that can generate diverse representations across multiple tasks, improving generalization and scalability in computer vision.

Contribution

It proposes a novel flow-matching approach using a universal velocity field to unify multiple vision tasks within a single model, inspired by large language models.

Findings

Achieves competitive performance in classification, detection, segmentation, depth estimation, and image-text retrieval.

Outperforms prior generalist and some specialist models in zero-shot and fine-tuned settings.

Demonstrates robustness, scalability, and strong generalization capabilities.

Abstract

Recent advances in diffusion models have achieved remarkable success in isolated computer vision tasks such as text-to-image generation, depth estimation, and optical flow. However, these models are often restricted by a ``single-task-single-model'' paradigm, severely limiting their generalizability and scalability in multi-task scenarios. Motivated by the cross-domain generalization ability of large language models, we propose a universal visual perception framework based on flow matching that can generate diverse visual representations across multiple tasks. Our approach formulates the process as a universal flow-matching problem from image patch tokens to task-specific representations rather than an independent generation or regression problem. By leveraging a strong self-supervised foundation model as the anchor and introducing a multi-scale, circular task embedding mechanism, our method learns a universal velocity field to bridge the gap between heterogeneous tasks, supporting efficient and flexible representation transfer. Extensive experiments on classification, detection, segmentation, depth estimation, and image-text retrieval demonstrate that our model achieves competitive performance in both zero-shot and fine-tuned settings, outperforming prior generalist and several specialist models. Ablation studies further validate the robustness, scalability, and generalization of our framework. Our work marks a significant step towards general-purpose visual perception, providing a solid foundation for future research in universal vision modeling.