Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Edit2Interp: Adapting Image Foundation Models from Spatial Editing to Video Frame Interpolation with Few-Shot Learning

Nasrin Rahimi, Mısra Yavuz, Burak Can Biner, Yunus Bilge Kurt, Ahmet Rasim Emirdağı, Süleyman Aslan, Görkay Aydemir, M. Akın Yılmaz, A. Murat Tekalp

Abstract

Pre-trained image editing models exhibit strong spatial reasoning and object-aware transformation capabilities acquired from billions of image-text pairs, yet they possess no explicit temporal modeling. This paper demonstrates that these spatial priors can be repurposed to unlock temporal synthesis capabilities through minimal adaptation - without introducing any video-specific architecture or motion estimation modules. We show that a large image editing model (Qwen-Image-Edit), originally designed solely for static instruction-based edits, can be adapted for Video Frame Interpolation (VFI) using only 64-256 training samples via Low-Rank Adaptation (LoRA). Our core contribution is revealing that the model's inherent understanding of "how objects transform" in static scenes contains latent temporal reasoning that can be activated through few-shot fine-tuning. While the baseline model completely fails at producing coherent intermediate frames, our parameter-efficient adaptation successfully unlocks its interpolation capability. Rather than competing with task-specific VFI methods trained from scratch on massive datasets, our work establishes that foundation image editing models possess untapped potential for temporal tasks, offering a data-efficient pathway for video synthesis in resource-constrained scenarios. This bridges the gap between image manipulation and video understanding, suggesting that spatial and temporal reasoning may be more intertwined in foundation models than previously recognized

Edit2Interp: Adapting Image Foundation Models from Spatial Editing to Video Frame Interpolation with Few-Shot Learning

Abstract

Pre-trained image editing models exhibit strong spatial reasoning and object-aware transformation capabilities acquired from billions of image-text pairs, yet they possess no explicit temporal modeling. This paper demonstrates that these spatial priors can be repurposed to unlock temporal synthesis capabilities through minimal adaptation - without introducing any video-specific architecture or motion estimation modules. We show that a large image editing model (Qwen-Image-Edit), originally designed solely for static instruction-based edits, can be adapted for Video Frame Interpolation (VFI) using only 64-256 training samples via Low-Rank Adaptation (LoRA). Our core contribution is revealing that the model's inherent understanding of "how objects transform" in static scenes contains latent temporal reasoning that can be activated through few-shot fine-tuning. While the baseline model completely fails at producing coherent intermediate frames, our parameter-efficient adaptation successfully unlocks its interpolation capability. Rather than competing with task-specific VFI methods trained from scratch on massive datasets, our work establishes that foundation image editing models possess untapped potential for temporal tasks, offering a data-efficient pathway for video synthesis in resource-constrained scenarios. This bridges the gap between image manipulation and video understanding, suggesting that spatial and temporal reasoning may be more intertwined in foundation models than previously recognized
Paper Structure (11 sections, 5 equations, 3 figures, 1 table)

This paper contains 11 sections, 5 equations, 3 figures, 1 table.

Table of Contents

  1. Introduction
  2. Related Works
  3. Method
  4. Foundation Model Backbone: Qwen-Image-Edit
  5. Repurposing for VFI via Constrained Editing and LoRA
  6. Experiments and Results
  7. Experimental Setup
  8. Quantitative Results
  9. Visual Results
  10. Ablation Studies
  11. Conclusion

Figures (3)

  • Figure 1: Conceptual comparison of VFI paradigms. (a) Traditional VFI networks,(b) Task-specific diffusion models, and (c) Our Edit2Interp approach
  • Figure 2: Overview of the Edit2Interp framework
  • Figure 3: Visual comparison across datasets. Top row for each dataset: ground truth frames $t\!-\!1$, $t$, $t\!+\!1$, baseline interpolation $t$, and LoRA interpolation $t$. Bottom row for each dataset: absolute difference maps with respect to the ground truth frame $t$.