Consistent and Editable: A Balanced Framework for Text-Guided Video Editing
2026-07-06 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors developed a new method called EquiEdit to improve video editing guided by text. Their method focuses on making the changes look smooth over time (temporal consistency) while still allowing lots of editing options (editability), which is usually hard to balance. They introduced a special tool (temporal Mamba module) that looks at video frames in multiple directions to keep the video consistent from frame to frame. They also use a noise strategy based on Fourier transforms to keep the video's original structure while editing. Tests show their method makes edited videos both consistent over time and faithful to the original.
diffusion modelsvideo editingtemporal consistencyeditabilitytemporal Mamba moduleFourier transformnoise injectionlatent noisetext-guided video editing
Authors
Tao Jin, Li Xiao
Abstract
Recently, diffusion models have achieved considerable success in the text-guided video editing domain. However, existing works often struggle to balance the trade-off between temporal consistency and editability in video editing, with consistency and editability typically being inversely related. To address this, we propose a high-quality video editing framework enhanced for consistency and editability, named EquiEdit, which improves coordinatively the temporal consistency and editability of the edited videos while achieving a balance between the two. In terms of temporal consistency, the proposed temporal Mamba module with a tailored temporal-aware scanning scans fused video sequences following four designed directions, effectively enhancing the inter-frame consistency of edited videos. For editability, we design a noise injection strategy based on the spectral transformation to increase editing flexibility, where the Fourier transform is used to preserve the hidden structure in the initial latent noise used for editing, ensuring inter-frame consistency of the edited video and fidelity to the input video. Extensive qualitative and quantitative experiments demonstrate the effectiveness of our method in terms of temporal consistency and editability, as well as its great fidelity to the input video itself.