Beyond Single Prompts: Synergistic Fusion and Arrangement for VICL

TL;DR

This paper introduces an end-to-end VICL framework that fuses multiple prompts and exploits their arrangements to improve visual inpainting tasks, addressing limitations of previous single-prompt methods.

Contribution

It proposes a novel adaptive fusion module and arrangement-specific lightweight MLPs, along with bidirectional fine-tuning, to enhance prompt utilization and model adaptability in VICL.

Findings

Outperforms existing methods on foreground segmentation, detection, and colorization.

Demonstrates strong cross-task generalization.

Achieves superior results with minimal additional model complexity.

Abstract

Vision In-Context Learning (VICL) enables inpainting models to quickly adapt to new visual tasks from only a few prompts. However, existing methods suffer from two key issues: (1) selecting only the most similar prompt discards complementary cues from other high-quality prompts; and (2) failing to exploit the structured information implied by different prompt arrangements. We propose an end-to-end VICL framework to overcome these limitations. Firstly, an adaptive Fusion Module aggregates critical patterns and annotations from multiple prompts to form more precise contextual prompts. Secondly, we introduce arrangement-specific lightweight MLPs to decouple layout priors from the core model, while minimally affecting the overall model. In addition, an bidirectional fine-tuning mechanism swaps the roles of query and prompt, encouraging the model to reconstruct the original prompt from fused context and thus enhancing collaboration between the fusion module and the inpainting model. Experiments on foreground segmentation, single-object detection, and image colorization demonstrate superior results and strong cross-task generalization of our method.