mlsanityData to 14 Oct 2026

Articulated Kinematics Distillation from Video Diffusion Models

arXiv 2504.01204 ↗First posted 1 Apr 2025cs.GR cs.CVcited by 13

On the map: Diffusion models → Video generation

Abstract (arXiv)

We present Articulated Kinematics Distillation (AKD), a framework for generating high-fidelity character animations by merging the strengths of skeleton-based animation and modern generative models. AKD uses a skeleton-based representation for rigged 3D assets, drastically reducing the Degrees of Freedom (DoFs) by focusing on joint-level control, which allows for efficient, consistent motion synthesis. Through Score Distillation Sampling (SDS) with pre-trained video diffusion models, AKD distills complex, articulated motions while maintaining structural integrity, overcoming challenges faced by 4D neural deformation fields in preserving shape consistency. This approach is naturally compatible with physics-based simulation, ensuring physically plausible interactions. Experiments show that AKD achieves superior 3D consistency and motion quality compared with existing works on text-to-4D generation. Project page: https://research.nvidia.com/labs/dir/akd/

Key facts. Cited by 13 papers in the map (data through 14 Oct 2026). Placed in Video generation, one of 14 directions in Diffusion models. Builds on 54 of its 67 references that are in the map. First posted on arXiv 1 Apr 2025.

Builds on

54 of 67 references.

Closest ancestors · 5

Same direction, newest first

  1. 2024CogVideoX: Text-to-Video Diffusion Models with An Expert Transformer12 Aug
  2. 2024VideoCrafter2: Overcoming Data Limitations for High-Quality Video Diffusion Models17 Jan
  3. 2023Single Motion Diffusion12 Feb
  4. 2022PhysDiff: Physics-Guided Human Motion Diffusion Model5 Dec
  5. 2022Human Motion Diffusion Model29 Sep

Other references · 49

  1. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  2. 2014Microsoft COCO: Common Objects in Contextcited by 17,608
  3. 20233D Gaussian Splatting for Real-Time Radiance Field Renderingcited by 6,226
  4. 2022Instant Neural Graphics Primitives with a Multiresolution Hash Encodingcited by 3,723
  5. 2022DreamFusion: Text-to-3D using 2D Diffusioncited by 3,025
  6. 2020D-NeRF: Neural Radiance Fields for Dynamic Scenescited by 1,465
  7. 2019AMASS: Archive of Motion Capture as Surface Shapescited by 1,236
  8. 2020Nerfies: Deformable Neural Radiance Fieldscited by 1,198
  9. 2017End-to-end Recovery of Human Shape and Posecited by 885
  10. 2016Training Deep Nets with Sublinear Memory Costcited by 848
  11. 2023Mip-Splatting: Alias-free 3D Gaussian Splattingcited by 667
  12. 2023MotionGPT: Human Motion as a Foreign Languagecited by 527
37 more
  1. 2023PhysGaussian: Physics-Integrated 3D Gaussians for Generative Dynamicscited by 377
  2. 2023Humans in 4D: Reconstructing and Tracking Humans with Transformerscited by 350
  3. 2024VideoPhy: Evaluating Physical Commonsense for Video Generationcited by 228
  4. 2023Text-To-4D Dynamic Scene Generationcited by 205
  5. 2021BANMo: Building Animatable 3D Neural Models from Many Casual Videoscited by 197
  6. 2023DreamGaussian4D: Generative 4D Gaussian Splattingcited by 194
  7. 2024PhysDreamer: Physics-Based Interaction with 3D Objects via Video Generationcited by 191
  8. 20234D-fy: Text-to-4D Generation Using Hybrid Score Distillation Samplingcited by 185
  9. 2024SplattingAvatar: Realistic Real-Time Human Avatars with Mesh-Embedded Gaussian Splattingcited by 172
  10. 2023Align Your Gaussians: Text-to-4D with Dynamic 3D Gaussians and Composed Diffusion Modelscited by 169
  11. 2023GauHuman: Articulated Gaussian Splatting from Monocular Human Videoscited by 168
  12. 2018Monocular Total Capture: Posing Face, Body, and Hands in the Wildcited by 162
  13. 2023HUGS: Human Gaussian Splatscited by 154
  14. 2024L4GM: Large 4D Gaussian Reconstruction Modelcited by 144
  15. 2024SV4D: Dynamic 3D Content Generation with Multi-Frame and Multi-View Consistencycited by 144
  16. 2023Consistent4D: Consistent 360deg Dynamic Object Generation from Monocular Videocited by 125
  17. 2023GART: Gaussian Articulated Template Modelscited by 121
  18. 2022MagicPony: Learning Articulated 3D Animals in the Wildcited by 112
  19. 2023Human Gaussian Splatting: Real-time Rendering of Animatable Avatarscited by 105
  20. 2024STAG4D: Spatial-Temporal Anchored Generative 4D Gaussianscited by 102
  21. 2023A Unified Approach for Text- and Image-guided 4D Scene Generationcited by 82
  22. 2024TC4D: Trajectory-Conditioned Text-to-4D Generationcited by 77
  23. 2022LASSIE: Learning Articulated Shapes from Sparse Image Ensemble via 3D Part Discoverycited by 66
  24. 2024SC4D: Sparse-Controlled Video-to-4D Generation and Motion Transfercited by 61
  25. 2024Learning the 3D Fauna of the Webcited by 59
  26. 2024Animate3D: Animating Any 3D Model with Multi-view Video Diffusioncited by 56
  27. 2023PIE-NeRF: Physics-based Interactive Elastodynamics with NeRFcited by 51
  28. 2024PhysAvatar: Learning the Physics of Dressed 3D Avatars from Visual Observationscited by 51
  29. 2023Reconstructing Animatable Categories from Videoscited by 47
  30. 2024NPGA: Neural Parametric Gaussian Avatarscited by 46
  31. 2023Farm3D: Learning Articulated 3D Animals by Distilling 2D Diffusioncited by 38
  32. 2022Hi-LASSIE: High-Fidelity Articulated Shape and Skeleton Discovery from Sparse Image Ensemblecited by 31
  33. 2024GVA: Reconstructing Vivid 3D Gaussian Avatars from Monocular Videoscited by 22
  34. 2023Dynamic Mesh-Aware Radiance Fieldscited by 21
  35. 2023Ponymation: Learning Articulated 3D Animal Motions from Unlabeled Online Videoscited by 21
  36. 2024DressRecon: Freeform 4D Human Reconstruction from Monocular Videocited by 12
  37. 2024Tetrahedron Splatting for 3D Generationcited by 9

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 2; 2026: 11.

13 papers.

Sort:

By area:

  1. 2025Puppeteer: Rig and Animate Your 3D Modelscited by 43
  2. 2025AnimaMimic: Imitating 3D Animation from Video Priorscited by 7
  3. 2026Preference Score Distillation: Leveraging 2D Rewards to Align Text-to-3D Generation with Human Preferencecited by 2
  4. 2026SV-GS: Sparse View 4D Reconstruction with Skeleton-Driven Gaussian Splattingcited by 1
  5. 2026Ani3DHuman: Photorealistic 3D Human Animation with Self-guided Stochastic Samplingcited by 1
  6. 2026Sketch2Motion: Text-driven 2D Sketch to 3D Animation via Diffusion-guided Skeleton Optimizationcited by 1
  7. 2026Follow Your Track: Precise Skeleton Animation Controlled by 3D Trajectoriescited by 1
  8. 2026ArtHOI: Articulated Human-Object Interaction Synthesis by 4D Reconstruction from Video Priorscited by 0
  9. 2026PhyGenHOI: Physically-Aware 4D Generation of Dynamic Human-Object Interactionscited by 0
  10. 2026Controllable Dynamic 3D Shape Generation via 3D Trajectories and Textcited by 0
  11. 2026AnimaSpark: A Feed-Forward Method for Animating Arbitrary 3D Objectscited by 0
  12. 2026G-Skin: Learning to Bind 3D Gaussians with Generative Visual Priorscited by 0
1 more
  1. 2026Strike a Chord! Modal Kinetic Typographycited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024Towards motion from video diffusion models
  2. 2025TriDiff-4D: Fast 4D Generation through Diffusion-based Triplane Re-posing
  3. 2025AnimaX: Animating the Inanimate in 3D with Joint Video-Pose Diffusion Models
  4. 2025Scalable Motion In-betweening via Diffusion and Physics-Based Character Adaptation
  5. 2026PWM-ArtGen: Part World Model for Articulated Object Generation
  6. 2025Learning to Generate Rigid Body Interactions with Video Diffusion Models
  7. 2023AnimatableDreamer: Text-Guided Non-rigid 3D Model Generation and Reconstruction with Canonical Score Distillation
  8. 2025FreeArt3D: Training-Free Articulated Object Generation using 3D Diffusion