Skip to main content
GameDev.net gamedev.net

3D Animation News

The latest 3D Animation coverage curated for game developers.

KM-Speaker: Keypoint-Based Style Control for High-Quality Speech-Driven 3D Facial Animation and Dialogue Localization

For teams doing facial animation or localization, this paper tackles a real production gap: getting both believable speech-driven motion and usable style control from a reference performance. KM-Speaker separates lip …

Research arXiv cs.GR · 3 months, 1 week ago
R-DMesh: Video-Guided 3D Animation via Rectified Dynamic Mesh Flow

A new paper tackles a very practical 3D animation failure mode: the source mesh and reference video often start from different poses. R-DMesh learns a rectification offset before animation, so …

Research arXiv cs.GR · 4 months, 3 weeks ago
FatigueFusion: Latent Space Fusion for Fatigue-Driven Motion Synthesis

FatigueFusion introduces a novel deep-learning architecture that allows developers to synthesize fatigue-driven motion in 3D animations. This tool offers a significant advancement for graphics programmers and animators, enabling the creation …

Research arXiv cs.GR · 5 months, 3 weeks ago
MATHDance: Mamba-Transformer Architecture with Uniform Tokenization for High-Quality 3D Dance Generation

The MATHDance framework introduces a novel approach to music-to-dance generation, enhancing choreographic consistency in 3D animations. This is particularly relevant for artists and programmers focused on integrating realistic dance movements …

Research arXiv cs.GR · 6 months, 1 week ago
Ani3DHuman: Photorealistic 3D Human Animation with Self-guided Stochastic Sampling

Ani3DHuman introduces a significant leap in 3D human animation by combining kinematics with video diffusion. This framework addresses the persistent challenges of photorealism and non-rigid dynamics, making it particularly relevant …

Research arXiv cs.GR · 7 months, 2 weeks ago
Skin Tokens: A Learned Compact Representation for Unified Autoregressive Rigging

The introduction of SkinTokens revolutionizes rigging in 3D animation, offering a 98%-133% improvement in skinning accuracy. This advancement is crucial for artists and programmers, streamlining workflows and enhancing fidelity in …

Research arXiv cs.GR · 8 months ago
PromptVFX: Text-Driven Fields for Open-World 3D Gaussian Animation

PromptVFX introduces a groundbreaking text-driven framework for 3D Gaussian animation, significantly reducing the complexity of creating visual effects. This innovation allows both novice and expert users to generate compelling animations …

Research arXiv cs.GR · 8 months, 1 week ago
Mocap Anywhere: Towards Pairwise-Distance based Motion Capture in the Wild (for the Wild)

A new motion capture system, Wild-Poser (WiP), leverages sparse pairwise distance measurements to reconstruct 3D motion without external cameras. This innovation allows for robust motion capture in uncontrolled outdoor environments, …

Research arXiv cs.GR · 8 months, 1 week ago
PartMotionEdit: Fine-Grained Text-Driven 3D Human Motion Editing via Part-Level Modulation

PartMotionEdit introduces a significant leap in 3D human motion editing by allowing fine-grained control over part-specific motions. This innovation is particularly relevant for animators and technical artists, as it enables …

Research arXiv cs.GR · 9 months, 1 week ago
PhysTalk: Language-driven Real-time Physics in 3D Gaussian Scenes

PhysTalk introduces a groundbreaking approach to real-time physics in 3D environments, enabling developers to create interactive animations from text prompts. This innovation significantly reduces the need for extensive rendering and …

Research arXiv cs.GR · 9 months, 1 week ago
Animus3D: Text-driven 3D Animation via Motion Score Distillation

Animus3D introduces a groundbreaking framework for generating 3D animations from text prompts, enhancing motion fidelity significantly. This tool is particularly relevant for artists and animators looking to streamline their workflow …

Research arXiv cs.GR · 9 months, 3 weeks ago
Towards Unified Co-Speech Gesture Generation via Hierarchical Implicit Periodicity Learning

The introduction of Hierarchical Implicit Periodicity (HIP) learning marks a significant advancement in generating realistic co-speech gestures for 3D avatars. This method addresses the limitations of existing models by effectively …

Research arXiv cs.GR · 9 months, 3 weeks ago
Learning to Control Physically-simulated 3D Characters via Generating and Mimicking 2D Motions

A new framework, Mimic2DM, allows developers to control 3D characters using only 2D motion data, significantly reducing reliance on costly 3D motion capture. This innovation is particularly beneficial for animators …

Research arXiv cs.GR · 10 months ago
Human Geometry Distribution for 3D Animation Generation

A new two-stage framework for generating realistic human geometry animations has been introduced, significantly improving quality with a 90% reduction in Chamfer Distance. This advancement is particularly relevant for graphics …

Research arXiv cs.GR · 10 months ago
PESTalk: Speech-Driven 3D Facial Animation with Personalized Emotional Styles

PESTalk introduces a groundbreaking method for creating 3D facial animations driven by speech, enhancing emotional expressiveness. This innovation is particularly relevant for artists and animators, as it allows for personalized …

Research arXiv cs.GR · 10 months ago