Our work "Towards Mocap Anywhere: Pairwise Distance-based Motion Capture in the Wild, for the Wild" has been accepted for publication in ACM Transactions on Graphics

29 Aug 2026

Wild-Poser (WiP) is a real-time Transformer-based system that reconstructs full-body 3D motion from sparse pairwise distance measurements using body-mounted UWB sensors, without relying on cameras or environmental infrastructure. WiP robustly predicts accurate 3D joint positions from noisy measurements and generalizes across diverse human and non-human morphologies in real-world settings. The paper has been accepted for publication in ACM Transactions on Graphics (TOG), with an Impact Factor of 13.0.

Our work "Cross-Morphology Motion Transfer with Semantic Style Alignment" has been accepted for publication in IEEE Transactions on Visualization and Computer Graphics

28 Aug 2026

We present a neural framework for cross-morphology motion transfer that enables realistic animation retargeting across characters with diverse skeletal structures. By combining morphology-agnostic control signals with morphology-specific latent codebooks, our method preserves both the semantic intent of the source motion and the natural motion style of the target character. The framework supports accurate and scalable all-to-all retargeting across bipeds and quadrupeds, with optional user control for fine-grained motion editing. The paper has been accepted for publication in IEEE Transactions on Visualization and Computer Graphics (TVCG), with an Impact Factor of 6.8.

Our work "MultiAct: Text-to-Motion Generation from Composite Text via Tailored Attention Guidance" has been accepted for publication at SIGGRAPH 2026

12 Jul 2026

MultiAct is an inference-time framework that enhances pretrained text-to-motion models to better capture multiple simultaneous actions by adaptively boosting attention to underrepresented prompt components. Without requiring retraining, it improves semantic completeness and action coverage, achieving stronger performance on complex multi-action prompts across quantitative and qualitative evaluations. The paper has been accepted for publication at SIGGRAPH 2026.

Our work "MotionPyramid: Controllable Motion Synthesis via Stylized Phase Manifolds" has been accepted at ACM SIGGRAPH / EG SCA

08 Jul 2026

MotionPyramid introduces stylized phase manifolds, a compact and interpretable latent representation that disentangles motion semantics, temporal dynamics, and style in an unsupervised, low-dimensional space. By using this representation as a structured bridge between text and motion, MotionPyramid enables fine-grained control over diffusion-based motion generation, producing high-quality, diverse motions with improved text-to-motion alignment. The paper has been accepted for presentation at the 25th ACM SIGGRAPH / Eurographics Symposium on Computer Animation (SCA 2026), Barcelona, Spain, and is published in Computer Graphics Forum, with IF: 2.96

Best Paper Award

31 Oct 2025

We are delighted to announce that the work of our students Marios Stylianou and Marios Demetriou, under the supervision of Dr. Andreas Aristidou, titled “Disaster Evacuation of the Old City of Nicosia”, has been awarded the Best Student Paper at the 8th International Disaster and Risk Conference (IDRC 2025).