Avatars News
The latest Avatars coverage curated for game developers.
arXiv cs.GR details ARS-Avatar, a surfel-based method for animatable, relightable human avatars from multi-view images under unknown lighting. The system adds deformation priors, deferred shading for BRDF estimation, and learnable …
UMA is a new human-avatar pipeline aimed at producing ultra-detailed digital people through multi-level surface alignment. Published in ACM Transactions on Graphics in February 2027, it targets higher-fidelity body representation …
A new data-driven motion model can reproduce an individual's repetitive movement signature from real human data. Built around LSTM networks, it captures both velocity distribution and amplitude envelopes, which could …
A new single-image avatar pipeline can generate photorealistic 3D head models with real-time expression control. S-Avatar combines diffusion-guided 3D Gaussian splatting with FLAME alignment to improve view consistency under motion …
A new face-reconstruction pipeline can build a fixed-topology mesh from 16 views in under 3 seconds on a single RTX 4090. UVFaceFusion swaps heuristic mesh cleanup for learnable fusion in …
AI-driven avatar tools are being pushed toward supporting limb differences, amputations, and other non-normative bodies in XR social spaces. The work highlights a real production gap: most avatar systems still …
A new avatar simulation approach tries to solve a long-standing problem: making human characters physically interact without collapsing their pose fidelity. PIAvatar decouples kinematic velocity from deformation gradient, then runs …
FiCA turns a single portrait into a drivable, photoreal 3D Gaussian avatar, cutting out per-person test-time optimization. The pipeline blends human-centric vision foundation models with diffusion and a mesh-refinement stage, …
AvatarMix tackles a real pain point in avatar outfit transfer: keeping identity and clothing intact without seam or intersection artifacts. It composes head and body from two high-fidelity Gaussian avatars, …
A pose-conditioned 3D Gaussian avatar can now carry a learned latent state that evolves over time, which helps with loose clothing and other motion that pose alone can't explain. The …
arXiv cs.GR details a causal, zero-lookahead system for personalized audio-driven facial motion that targets real-time streaming. The method uses a multi-resolution motion tokenizer and a multi-modal style retriever to adapt …
The emergence of AI-generated anime avatars is reshaping the landscape of digital character creation, particularly for graphics programmers and artists. This technology leverages deep learning to automate the generation of …
The introduction of Retrieval-Augmented Faces (RAF) offers a significant leap in animatable head avatar technology, enhancing expression generalization without the need for complex templates. This method allows for a broader …
OFERA introduces a groundbreaking method for controlling photorealistic avatars in VR using blendshape signals from headsets. This approach eliminates the need for additional sensors or cameras, addressing comfort and privacy …
Netflix's acquisition of Ready Player Me signals a significant shift in the gaming landscape, particularly for developers focused on avatar creation and customization. This move could enhance Netflix's gaming offerings, …
The introduction of Gaussian Pixel Codec Avatars (GPiCA) could revolutionize how we create and render photorealistic avatars in games. By combining triangle meshes with 3D Gaussians, this hybrid approach enhances …