DynaDrag: Dynamic Drag-Style Image Editing by Motion Prediction

2026-01-02

arXiv.org

AI Summary

This work proposes DynaDrag, the first drag-based image editing method built upon a “predict-and-move” framework, addressing the limitations of traditional approaches that often suffer from tracking failures or domain mismatches between source and target regions. By iteratively performing motion prediction and motion supervision while dynamically adjusting active control points, DynaDrag achieves high-quality, pixel-level editing without relying on conventional tracking mechanisms. The introduction of a dynamic control point strategy significantly enhances both the plausibility and controllability of edits. Extensive experiments on face and human datasets demonstrate that DynaDrag consistently outperforms existing methods in terms of visual naturalness and structural consistency.

Technology Category

• Computer Vision: Motion & Tracking
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Intelligent Robots: Motion and Path Planning

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

To achieve pixel-level image manipulation, drag-style image editing which edits images using points or trajectories as conditions is attracting widespread attention. Most previous methods follow move-and-track framework, in which miss tracking and ambiguous tracking are unavoidable challenging issues. Other methods under different frameworks suffer from various problems like the huge gap between source image and target edited image as well as unreasonable intermediate point which can lead to low editability. To avoid these problems, we propose DynaDrag, the first dragging method under predict-and-move framework. In DynaDrag, Motion Prediction and Motion Supervision are performed iteratively. In each iteration, Motion Prediction first predicts where the handle points should move, and then Motion Supervision drags them accordingly. We also propose to dynamically adjust the valid handle points to further improve the performance. Experiments on face and human datasets showcase the superiority over previous works.

Problem

• drag-style image editing
• motion prediction
• pixel-level manipulation
• tracking ambiguity
• image editability

Innovation

• drag-style editing
• motion prediction
• predict-and-move framework
• dynamic handle adjustment
• image manipulation