Video Style Transfer¶
Video
Open Source¶
-
DreamStyle Discovered by AI
2026
A unified video stylization framework that supports multiple style conditions, including text prompts, style reference images, and stylized first frames, to achieve consistent and flexible video-to-video style transfer.
diffusionstyle-transfervideo-to-video -
EchoStyle: Unlocking High-Fidelity Video Stylization with Reverse Data Synthesis Discovered by AI
2026
A high-fidelity video stylization framework built on the Wan2.2 video generation model that uses a reverse data synthesis pipeline and an init-follow-mode mechanism to maintain temporal and motion coherence across arbitrary video lengths.
diffusionstyle-transferwan2.2 -
VISTA: Triplet-Supervised Video Style Transfer with Diffusion Transformers Discovered by AI
2026
A video style transfer framework built on Diffusion Transformers that uses triplet supervision to achieve high temporal consistency and style faithfulness without relying on brittle propagation heuristics.
diffusion-transformertemporal-consistencyvideo-style-transfer -
FreeViS: Training-free Video Stylization with Inconsistent References Discovered by AI
2025
A training-free video stylization framework that leverages multiple inconsistent stylized references to generate high-quality stylized videos with rich style details and strong temporal coherence.
diffusiontraining-freevideo-style-transfer -
FRESCO-v2 Discovered by AI
2025
An upgraded zero-shot video translation and editing framework that leverages frame spatial-temporal correspondence to enforce temporal consistency without requiring model training.
style-transfertemporal-consistencyzero-shot -
PickStyle Discovered by AI
2025
A video-to-video style transfer framework that augments pretrained video diffusion backbones with context-style adapters, enabling precise style application while preserving the original video's motion and context.
adapterdiffusion-modelstyle-transfer -
TRAC Discovered by AI
2025
A video style transfer method that introduces Trajectory Reset Attention Control (TRAC) and content-style bridging to prevent content leakage and style misalignment during image-driven style transfer.
attention-controldiffusionstyle-transfer -
V-Stylist Discovered by AI
2025
A multi-agent video stylization system driven by multimodal large language models (MLLMs) that decomposes input videos into scenes, generates style descriptions, and coordinates stylization to handle complex transitions.
mllmmulti-agentstyle-transfer -
FRESCO: Spatial-Temporal Correspondence for Zero-Shot Video Translation Discovered by AI
2024
A zero-shot video translation framework that enforces spatial-temporal correspondence (both intra-frame and inter-frame) to ensure temporal coherence and prevent flickering in video style transfer.
diffusiontemporal-consistencyvideo-style-transfer -
NijiGAN: Transform What You See into Anime with Contrastive Semi-Supervised Learning and Neural Ordinary Differential Equations Discovered by AI
2024
An efficient real-to-anime image and video translation framework that combines Neural Ordinary Differential Equations (NeuralODEs) and contrastive semi-supervised learning. It achieves high-fidelity anime scene translation with significantly fewer parameters than previous models.
animeganneural-odevideo-style-transfer -
RAIN: Real-time Animation of Infinite Video Stream Discovered by AI
2024
A real-time, low-latency pipeline for animating infinite video streams (video-to-video style transfer) on a single consumer GPU. It uses a sliding-window frame-token attention mechanism and a keyframe-guided denoising strategy to maintain temporal consistency and visual quality without memory constraints.
diffusionreal-timevideo-style-transfervideo-to-video -
StreamV2V Discovered by AI
2024
A real-time streaming video-to-video translation framework that utilizes a backward-looking feature bank to maintain temporal consistency across unlimited frames.
diffusionreal-timevideo-to-video -
StyleMaster: Stylize Your Video with Artistic Generation and Translation Discovered by AI
2024
A video stylization framework that integrates a motion adapter and a gray tile ControlNet to enable precise content guidance and local texture extraction, enhancing dynamic quality and style resemblance for video generation and translation.
controlnetstyle-transfervideo-generation -
UniVST Discovered by AI
2024
A training-free localized video style transfer framework based on diffusion models, featuring a point-matching mask propagation strategy and a style injection module to apply styles to specific regions while maintaining temporal coherence.
localized-transferstyle-transfertraining-free -
BIVDiff: A Training-Free Framework for General-Purpose Video Synthesis via Bridging Image and Video Diffusion Models Discovered by AI
2023
BIVDiff is a training-free framework that bridges image and video diffusion models to enable diverse video-to-video synthesis and style editing with high temporal stability and strong controllability.
diffusiontraining-freevideo-style-transfervideo-to-video -
CoDeF Discovered by AI
2023
Content Deformation Fields (CoDeF) represents videos as a canonical content field and a temporal deformation field, enabling highly temporally consistent video style transfer and editing.
style-transfertemporal-consistencyvideo-editing -
Control-A-Video: Controllable Text-to-Video Generation with Diffusion Models Discovered by AI
2023
A controllable text-to-video diffusion model that generates temporally consistent videos conditioned on a sequence of control signals such as edge or depth maps, enabling precise video-to-video style transfer and editing.
controlnetdiffusionstyle-transfer -
FastBlend: a Powerful Model-Free Toolkit Making Video Stylization Easier Discovered by AI
2023
A model-free post-processing toolkit designed to eliminate flicker and enhance temporal consistency in video stylization workflows. It can be seamlessly integrated with diffusion models and ComfyUI/WebUI pipelines to produce smooth video-to-video style transfer results.
comfyuipost-processingtemporal-consistencyvideo-style-transfer -
FlowVid: Taming Imperfect Optical Flows for Consistent Video-to-Video Synthesis Discovered by AI
2023
A video-to-video synthesis framework that leverages spatial conditions and optical flow clues to maintain temporal consistency in video style transfer.
optical-flowtemporal-consistencyvideo-style-transfer -
LatentWarp: Consistent Diffusion Latents for Zero-Shot Video-to-Video Translation Discovered by AI
2023
LatentWarp enforces temporal consistency in zero-shot diffusion video-to-video translation by warping and aligning latent representations across adjacent frames using optical flow.
diffusionvideo-style-transfervideo-to-videozero-shot -
Rerender A Video Discovered by AI
2023
A zero-shot text-guided video-to-video translation framework that adapts image diffusion models to videos using key frame translation and temporal consistency constraints.
diffusionvideo-to-videozero-shot -
Style-A-Video: Agile Diffusion for Arbitrary Text-based Video Style Transfer Discovered by AI
2023
A zero-shot video stylization method that leverages a pretrained image latent diffusion model and a generative pre-trained transformer to achieve temporally consistent, text-guided video style transfer. It introduces a sampling strategy that preserves content structure while adapting to arbitrary artistic styles.
diffusiontext-guidedvideo-style-transferzero-shot -
StyleCrafter: Enhancing Stylized Text-to-Video Generation with Style Adapter Discovered by AI
2023
StyleCrafter enhances pre-trained text-to-video diffusion models with a lightweight style adapter, allowing reference image-guided video style transfer and stylized text-to-video generation without compromising video quality.
diffusionstyle-adaptervideo-generationvideo-style-transfer -
VToonify Discovered by AI
2022
A controllable high-resolution portrait video style transfer framework that leverages StyleGAN to generate high-quality artistic portrait videos with temporal consistency.
portraitstyle-transferstylegan -
EbSynth Discovered by AI
2019
A classic, highly efficient example-based image synthesis and style transfer tool that propagates hand-painted keyframe styles across video sequences.
keyframe-propagationstyle-transfervfx
Proprietary / Industry Tools¶
-
GoEnhance AI Discovered by AI
2024
A commercial AI video style transfer platform that allows users to transform existing footage into anime, claymation, Pixar-like, and other artistic styles with smooth motion tracking.
animeclaymationsaasvideo-style-transfer -
Mago Studio Discovered by AI
2024
A studio-grade AI rendering and video-to-video style transfer platform designed for animation and VFX pipelines. It integrates into existing toolchains to transform playblasts, previs, and live-action plates into final stylized renders while maintaining character integrity and temporal consistency.
animationrenderingstyle-transfervfx -
DomoAI Discovered by AI
2023
A popular commercial AI video generation and editing platform specializing in high-quality video-to-video style transfer. It allows creators to transform real-world footage into various artistic styles, including Japanese anime, 3D cartoon, and sketch, while maintaining motion continuity and lip-syncing.
animecommercialvideo-style-transfervideo-to-video -
Kaiber Discovered by AI
2023
A popular AI creative platform that offers video-to-video style transfer, audio-reactive animation, and text-to-video generation, widely used by musicians and creators for stylized video production.
audio-reactivestyle-transfervideo-to-video -
EbSynth Discovered by AI
A keyframe-based video style transfer tool that propagates the style of one or more hand-painted keyframes across an entire video sequence. It is widely used by artists and animators to create hand-painted, stylized, or cartoon animations from live-action footage.
animationkeyframestyle-transfer
