Delta-Adapter：基於範例的可擴展圖像編輯方法，採用單對監督

Delta-Adapter: Scalable Exemplar-Based Image Editing with Single-Pair Supervision

May 8, 2026

作者: Jiacheng Chen, Songze Li, Han Fu, Baoquan Zhao, Wei Liu, Yanyan Liang, Li Qing, Xudong Mao

cs.AI

摘要

基於範例的圖像編輯藉由一組來源-目標圖像對定義的轉換，將其應用於新的查詢圖像。現有方法依賴於「成對配對」的監督模式，需要兩組共享相同編輯語意的圖像對來學習目標轉換。此限制使得訓練資料難以大規模整理，並限制了在各種編輯類型間的泛化能力。我們提出Delta-Adapter，這是一種在單一配對監督下學習可遷移編輯語意的方法，無需文字引導。此方法不直接將範例對暴露給模型，而是利用預訓練的視覺編碼器來萃取編碼兩張圖像間視覺轉換的「語意差量」。此語意差量透過基於感知器的適配器注入到預訓練的圖像編輯模型中。由於目標圖像從未直接可見於模型，因此可將其作為預測目標，在無需額外範例對的情況下實現單一配對監督。此公式使我們能夠利用現有大規模編輯資料集進行訓練。為進一步促進忠實的轉換遷移，我們引入語意差量一致性損失函數，使生成輸出的語意變化與從範例對中萃取的真實語意差量對齊。大量實驗證明，Delta-Adapter在四種強基線方法上，針對已見編輯任務持續提升編輯準確性與內容一致性，同時在未見編輯任務上展現更有效的泛化能力。程式碼將於 https://delta-adapter.github.io 提供。

English

Exemplar-based image editing applies a transformation defined by a source-target image pair to a new query image. Existing methods rely on a pair-of-pairs supervision paradigm, requiring two image pairs sharing the same edit semantics to learn the target transformation. This constraint makes training data difficult to curate at scale and limits generalization across diverse edit types. We propose Delta-Adapter, a method that learns transferable editing semantics under single-pair supervision, requiring no textual guidance. Rather than directly exposing the exemplar pair to the model, we leverage a pre-trained vision encoder to extract a semantic delta that encodes the visual transformation between the two images. This semantic delta is injected into a pre-trained image editing model via a Perceiver-based adapter. Since the target image is never directly visible to the model, it can serve as the prediction target, enabling single-pair supervision without requiring additional exemplar pairs. This formulation allows us to leverage existing large-scale editing datasets for training. To further promote faithful transformation transfer, we introduce a semantic delta consistency loss that aligns the semantic change of the generated output with the ground-truth semantic delta extracted from the exemplar pair. Extensive experiments demonstrate that Delta-Adapter consistently improves both editing accuracy and content consistency over four strong baselines on seen editing tasks, while also generalizing more effectively to unseen editing tasks. Code will be available at https://delta-adapter.github.io.