pith. sign in

arxiv: 2504.14977 · v1 · pith:OZO4CRT5new · submitted 2025-04-21 · 💻 cs.CV

RealisDance-DiT: Simple yet Strong Baseline towards Controllable Character Animation in the Wild

classification 💻 cs.CV
keywords modelfoundationanimationcharactercontrollabledatasetrealisdance-ditbaseline
0
0 comments X
read the original abstract

Controllable character animation remains a challenging problem, particularly in handling rare poses, stylized characters, character-object interactions, complex illumination, and dynamic scenes. To tackle these issues, prior work has largely focused on injecting pose and appearance guidance via elaborate bypass networks, but often struggles to generalize to open-world scenarios. In this paper, we propose a new perspective that, as long as the foundation model is powerful enough, straightforward model modifications with flexible fine-tuning strategies can largely address the above challenges, taking a step towards controllable character animation in the wild. Specifically, we introduce RealisDance-DiT, built upon the Wan-2.1 video foundation model. Our sufficient analysis reveals that the widely adopted Reference Net design is suboptimal for large-scale DiT models. Instead, we demonstrate that minimal modifications to the foundation model architecture yield a surprisingly strong baseline. We further propose the low-noise warmup and "large batches and small iterations" strategies to accelerate model convergence during fine-tuning while maximally preserving the priors of the foundation model. In addition, we introduce a new test dataset that captures diverse real-world challenges, complementing existing benchmarks such as TikTok dataset and UBC fashion video dataset, to comprehensively evaluate the proposed method. Extensive experiments show that RealisDance-DiT outperforms existing methods by a large margin.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 8 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. CoMoVi: Co-Generation of 3D Human Motions and Realistic Videos

    cs.CV 2026-01 unverdicted novelty 7.0

    CoMoVi co-generates 3D human motions and 2D videos synchronously in a single diffusion denoising loop using 3D-to-2D projection and dual-branch diffusion with 3D-2D cross attentions.

  2. 3D Scene-Adaptive Trajectory-Controllable Human Image Animation with Camera Movement

    cs.CV 2026-06 unverdicted novelty 6.0

    Presents a scene-adaptive 3D human animation method using ground-adaptive motion retargeting and viewpoint-adaptive latent fusion to control human trajectories and camera views, reporting gains on two benchmarks.

  3. EMOSH: Expressive Motion and Shape Disentanglement for Human Animation

    cs.CV 2026-06 unverdicted novelty 6.0

    EMOSH proposes an Expressive Human Model with disentangled parameters, coarse-to-fine motion injection, and spatially-aligned conditioning to generate high-fidelity expressive human videos without driving-subject shap...

  4. Towards 3D-Aware Video Diffusion Models: Render-Free Human Motion Control with Mesh Tokenization

    cs.CV 2026-06 unverdicted novelty 6.0

    Introduces mesh tokenization to condition DiT-based video diffusion models directly on 3D human meshes for motion control without 2D rendering.

  5. EverAnimate: Minute-Scale Human Animation via Latent Flow Restoration

    cs.CV 2026-05 unverdicted novelty 6.0

    EverAnimate restores drifted latent flow trajectories in chunked video generation via persistent latent propagation and restorative flow matching, achieving measurable gains in PSNR, SSIM, LPIPS, and FID over prior lo...

  6. SteadyDancer: Harmonized and Coherent Human Image Animation with First-Frame Preservation

    cs.CV 2025-11 unverdicted novelty 6.0

    SteadyDancer is an I2V framework using condition reconciliation, synergistic pose modulation, and staged training to achieve robust first-frame preservation and coherent motion control in human image animation.

  7. 3D Scene-Adaptive Trajectory-Controllable Human Image Animation with Camera Movement

    cs.CV 2026-06 unverdicted novelty 5.0

    Presents a scene-adaptive 3D human image animation framework using ground-adaptive motion retargeting and viewpoint-adaptive latent fusion to control human and camera trajectories, claiming improvements on two benchmarks.

  8. Directing the World: Fast Autoregressive Video Generation with Compositional Human-Camera Control

    cs.CV 2026-06 unverdicted novelty 5.0

    A decoupled-control autoregressive video model using Fast-Slow Memory training, dynamic projection, and staged camera control to produce stable long-horizon outputs with human and viewpoint guidance.