【论文合集】Awesome Video Diffusion

A curated list of recent diffusion models for video generation, editing, restoration, understanding, nerf, etc.

来源：https://github.com/showlab/Awesome-Video-Diffusion#

目录

Open-source Toolboxes and Foundation Models

Video Generation

Video Editing

Long-form Video Generation and Completion

Human or Subject Motion

Video Enhancement and Restoration

3D / NeRF

Video Understanding

Healthcare and Biology

---

Open-source Toolboxes and Foundation Models

• I2VGen-XL (image-to-video / video-to-video)

• text-to-video-synthesis-colab

• VideoCrafter: A Toolkit for Text-to-Video Generation and Editing

• ModelScope (Text-to-video synthesis)

• Diffusers (Text-to-video synthesis)

Video Generation

• Hierarchical Masked 3D Diffusion Model for Video Outpainting (Sep., 2023)

• VideoGen: A Reference-Guided Latent Diffusion Approach for High Definition Text-to-Video Generation (Sep., 2023)

• MagicAvatar: Multimodal Avatar Generation and Animation (Aug., 2023)

• Empowering Dynamics-aware Text-to-Video Diffusion with Large Language Models (Aug., 2023)

• ModelScope Text-to-Video Technical Report (Aug., 2023)

• Dual-Stream Diffusion Net for Text-to-Video Generation (Aug., 2023)

• DragNUWA: Fine-grained Control in Video Generation by Integrating Text, Image, and Trajectory (Aug., 2023)

• InternVid: A Large-scale Video-Text Dataset for Multimodal Understanding and Generation (Jul., 2023)

• Animate-A-Story: Storytelling with Retrieval-Augmented Video Generation (Jul., 2023)

• AnimateDiff: Animate Your Personalized Text-to-Image Diffusion Models without Specific Tuning (Jul., 2023)

• DisCo: Disentangled Control for Referring Human Dance Generation in Real World (Jul., 2023)

• VideoComposer: Compositional Video Synthesis with Motion Controllability (Jun., 2023)

• Probabilistic Adaptation of Text-to-Video Models (Jun., 2023)

• Make-Your-Video: Customized Video Generation Using Textual and Structural Guidance (Jun., 2023)

• Gen-L-Video: Multi-Text to Long Video Generation via Temporal Co-Denoising (May, 2023)

• Control-A-Video: Controllable Text-to-Video Generation with Diffusion Models (May, 2023)

• ControlVideo: Training-free Controllable Text-to-Video Generation (May, 2023)

• Cinematic Mindscapes: High-quality Video Reconstruction from Brain Activity (May, 2023)

• Any-to-Any Generation via Composable Diffusion (May, 2023)

• VideoFactory: Swap Attention in Spatiotemporal Diffusions for Text-to-Video Generation (May, 2023)

• Preserve Your Own Correlation: A Noise Prior for Video Diffusion Models (May, 2023)

• Motion-Conditioned Diffusion Model for Controllable Video Synthesis (Apr., 2023)

• LaMD: Latent Motion Diffusion for Video Generation (Apr., 2023)

• Align your Latents: High-Resolution Video Synthesis with Latent Diffusion Models (CVPR 2023)

• Text2Performer: Text-Driven Human Video Generation (Apr., 2023)

• Generative Disco: Text-to-Video Generation for Music Visualization (Apr., 2023)

• Latent-Shift: Latent Diffusion with Temporal Shift (Apr., 2023)

• DreamPose: Fashion Image-to-Video Synthesis via Stable Diffusion (Apr., 2023)

• Follow Your Pose: Pose-Guided Text-to-Video Generation using Pose-Free Videos (Apr., 2023)

• Physics-Driven Diffusion Models for Impact Sound Synthesis from Videos (CVPR 2023)

• Seer: Language Instructed Video Prediction with Latent Diffusion Models (Mar., 2023)

• Text2video-Zero: Text-to-Image Diffusion Models Are Zero-Shot Video Generators (Mar., 2023)

• Conditional Image-to-Video Generation with Latent Flow Diffusion Models (CVPR 2023)

• Decomposed Diffusion Models for High-Quality Video Generation (CVPR 2023)

• Video Probabilistic Diffusion Models in Projected Latent Space (CVPR 2023)

• Learning 3D Photography Videos via Self-supervised Diffusion on Single Images (Feb., 2023)

• Structure and Content-Guided Video Synthesis With Diffusion Models (Feb., 2023)

• Tune-A-Video: One-Shot Tuning of Image Diffusion Models for Text-to-Video Generation (Dec., 2022)

• Mm-Diffusion: Learning Multi-Modal Diffusion Models for Joint Audio and Video Generation (CVPR 2023)

• Magvit: Masked Generative Video Transformer (Dec., 2022)

• VIDM: Video Implicit Diffusion Models (AAAI 2023)

• Latent Video Diffusion Models for High-Fidelity Video Generation With Arbitrary Lengths (Nov., 2022)

• SinFusion: Training Diffusion Models on a Single Image or Video (Nov., 2022)

• MagicVideo: Efficient Video Generation With Latent Diffusion Models (Nov., 2022)

• Imagen Video: High Definition Video Generation With Diffusion Models (Oct., 2022)

• Make-A-Video: Text-to-Video Generation without Text-Video Data (ICLR 2023)

• Diffusion Models for Video Prediction and Infilling (TMLR 2022)

• McVd: Masked Conditional Video Diffusion for Prediction, Generation, and Interpolation (NeurIPS 2022)

• Video Diffusion Models (Apr., 2022）

• Diffusion Probabilistic Modeling for Video Generation (Mar., 2022）

Video Editing

• MagicProp: Diffusion-based Video Editing via Motion-aware Appearance Propagation (Sep., 2023)

• MagicEdit: High-Fidelity and Temporally Coherent Video Editing (Aug., 2023)

• StableVideo: Text-driven Consistency-aware Diffusion Video Editing (ICCV 2023)

• CoDeF: Content Deformation Fields for Temporally Consistent Video Processing (Aug., 2023)

• TokenFlow: Consistent Diffusion Features for Consistent Video Editing (Jul., 2023)

• INVE: Interactive Neural Video Editing (Jul., 2023)

• VidEdit: Zero-Shot and Spatially Aware Text-Driven Video Editing (Jun., 2023)

• Rerender A Video: Zero-Shot Text-Guided Video-to-Video Translation (Jun., 2023)

• ControlVideo: Adding Conditional Control for One Shot Text-to-Video Editing (May, 2023)

• Make-A-Protagonist: Generic Video Editing with An Ensemble of Experts (May, 2023)

• Soundini: Sound-Guided Diffusion for Natural Video Editing (Apr., 2023)

• Zero-Shot Video Editing Using Off-the-Shelf Image Diffusion Models (Mar., 2023)

• Edit-A-Video: Single Video Editing with Object-Aware Consistency (Mar., 2023)

• FateZero: Fusing Attentions for Zero-shot Text-based Video Editing (Mar., 2023)

• Pix2video: Video Editing Using Image Diffusion (Mar., 2023)

• Video-P2P: Video Editing with Cross-attention Control (Mar., 2023)

• Dreamix: Video Diffusion Models Are General Video Editors (Feb., 2023)

• Shape-Aware Text-Driven Layered Video Editing (Jan., 2023)

• Speech Driven Video Editing via an Audio-Conditioned Diffusion Model (Jan., 2023)

• Diffusion Video Autoencoders: Toward Temporally Consistent Face Video Editing via Disentangled Video Encoding (CVPR 2023)

Long-form Video Generation and Completion

• MCVD: Masked Conditional Video Diffusion for Prediction, Generation, and Interpolation (NeurIPS 2022)

• NUWA-XL: Diffusion over Diffusion for eXtremely Long Video Generation (Mar., 2023)

• Flexible Diffusion Modeling of Long Videos (May, 2022)

Human or Subject Motion

• Avatars Grow Legs: Generating Smooth Human Motion from Sparse Tracking Inputs with Diffusion Model (CVPR 2023)

• InterGen: Diffusion-based Multi-human Motion Generation under Complex Interactions (Apr., 2023)

• ReMoDiffuse: Retrieval-Augmented Motion Diffusion Model (Apr., 2023)

• Human Motion Diffusion as a Generative Prior (Mar., 2023)

• Can We Use Diffusion Probabilistic Models for 3d Motion Prediction? (Feb., 2023)

• Single Motion Diffusion (Feb., 2023)

• HumanMAC: Masked Motion Completion for Human Motion Prediction (Feb., 2023）

• DiffMotion: Speech-Driven Gesture Synthesis Using Denoising Diffusion Model (Jan., 2023)

• Modiff: Action-Conditioned 3d Motion Generation With Denoising Diffusion Probabilistic Models (Jan., 2023)

• Unifying Human Motion Synthesis and Style Transfer With Denoising Diffusion Probabilistic Models (GRAPP 2023)

• Executing Your Commands via Motion Diffusion in Latent Space (CVPR 2023)

• Pretrained Diffusion Models for Unified Human Motion Synthesis (Dec., 2022)

• PhysDiff: Physics-Guided Human Motion Diffusion Model (Dec., 2022)

• BeLFusion: Latent Diffusion for Behavior-Driven Human Motion Prediction (Dec., 2022)

• Listen, Denoise, Action! Audio-Driven Motion Synthesis With Diffusion Models (Nov. 2022)

• Diffusion Motion: Generate Text-Guided 3d Human Motion by Diffusion Model (ICASSP 2023)

• Human Joint Kinematics Diffusion-Refinement for Stochastic Motion Prediction (Oct., 2022)

• Human Motion Diffusion Model (ICLR 2023)

• FLAME: Free-form Language-based Motion Synthesis & Editing (AAAI 2023)

• MotionDiffuse: Text-Driven Human Motion Generation with Diffusion Model (Aug., 2022)

• Stochastic Trajectory Prediction via Motion Indeterminacy Diffusion (CVPR 2022)

Video Enhancement and Restoration

• LDMVFI: Video Frame Interpolation with Latent Diffusion Models (Mar., 2023)

• CaDM: Codec-aware Diffusion Modeling for Neural-enhanced Video Streaming (Nov., 2022)

3D / NeRF

• Text2NeRF: Text-Driven 3D Scene Generation with Neural Radiance Fields (May, 2023)

• RoomDreamer: Text-Driven 3D Indoor Scene Synthesis with Coherent Geometry and Texture (May, 2023)

• NeuralField-LDM: Scene Generation with Hierarchical Latent Diffusion Models (CVPR 2023)

• Single-Stage Diffusion NeRF: A Unified Approach to 3D Generation and Reconstruction (Apr., 2023

• Instruct-NeRF2NeRF: Editing 3D Scenes with Instructions (Mar., 2023)

• DiffusioNeRF: Regularizing Neural Radiance Fields with Denoising Diffusion Models (Feb., 2023)

• NerfDiff: Single-image View Synthesis with NeRF-guided Distillation from 3D-aware Diffusion (Feb., 2023)

• DiffRF: Rendering-guided 3D Radiance Field Diffusion (CVPR 2023)

Video Understanding

• Exploring Diffusion Models for Unsupervised Video Anomaly Detection (Apr., 2023)

• PDPP:Projected Diffusion for Procedure Planning in Instructional Videos (CVPR 2023) 

• DiffTAD: Temporal Action Detection with Proposal Denoising Diffusion (Mar., 2023)

• DiffusionRet: Generative Text-Video Retrieval with Diffusion Model (Mar., 2023)

• Refined Semantic Enhancement Towards Frequency Diffusion for Video Captioning (Nov., 2022)

• A Generalist Framework for Panoptic Segmentation of Images and Videos (Oct., 2022)

Healthcare and Biology

• Annealed Score-Based Diffusion Model for Mr Motion Artifact Reduction (Jan., 2023)

• Feature-Conditioned Cascaded Video Diffusion Models for Precise Echocardiogram Synthesis (Mar., 2023)

• Neural Cell Video Synthesis via Optical-Flow Diffusion (Dec., 2022)