UniEdit: A Unified Tuning-Free Framework for Efficient Video Motion and Appearance Editing
UniEdit is a tuning-free framework that supports both video motion editing (e.g., changing the action from playing guitar to waving) and various video appearance editing scenarios (e.g., stylization, object replacement, background modification) by harnessing the power of a pre-trained text-to-video generator within an inversion-then-generation pipeline.
CCEdit: Versatile Generative Video Editing with Precise Control over Structure and Appearance
CCEdit is a versatile generative video editing framework that decouples structure and appearance control, enabling precise and creative editing capabilities through a novel trident network architecture.
Video Editing via Interpolative Non-autoregressive Masked Transformers
MaskINT, an efficient prompt-based video editing framework, disentangles the task into keyframes joint editing and structure-aware frame interpolation, eliminating the need for paired text-video datasets and significantly accelerating the processing time compared to diffusion-based methods.
VidEdit: Zero-Shot and Spatially Aware Text-Driven Video Editing
VidEdit is a novel method for zero-shot text-based video editing that guarantees robust temporal and spatial consistency by combining an atlas-based video representation with a pre-trained text-to-image diffusion model.
DragVideo: An Intuitive and Accurate Drag-Style Video Editing Framework
DragVideo proposes a novel framework that enables intuitive and accurate drag-style editing of videos while preserving spatio-temporal consistency.
FastVideoEdit: Leveraging Consistency Models for Efficient Text-to-Video Editing
FastVideoEdit proposes an efficient zero-shot video editing approach inspired by Consistency Models, reducing editing time while maintaining high quality.
RealCraft: Zero-Shot Video Editing with Attention Control
RealCraft ermöglicht formgerechtes, konsistentes Video-Editing ohne zusätzliche Parameter.
RealCraft: Zero-Shot Video Editing with Attention Control Method
RealCraft proposes an attention-control-based method for zero-shot real-world video editing, achieving shape-wise edits and enhanced temporal consistency without additional information.
Universal Representation Learning for Video Editing Components
Learning visual representations of editing components is crucial for video creation tasks, supported by a novel dataset and method.
VIDEOSHOP: Localized Semantic Video Editing with Noise-Extrapolated Diffusion Inversion
VIDEOSHOP enables precise video editing by propagating semantic changes across frames.
