Sinan Mutlu
Research Subjects
Research Interests
Sinan Mutlu has a total of 5 co-authors. Here are the top co-authors by number of shared publications:
Anastasios Drosou
Information Technologies Institute CERTH
1 shared publication- Georgios-Fotios Angelis
Information Technologies Institute, Centre for Research and Technology Hellas, Thessaloniki, Greece
1 shared publication Mete Ozay
Samsung R&D Institute UK (SRUK)
1 shared publicationPaul Wisbey
Samsung R&D Institute UK (SRUK)
1 shared publication- Savas Ozkan
Samsung R&D Institute UK (SRUK)
1 shared publication
Latest Publications
- Open Access
Mem-MLP: Real-Time 3D Human Motion Generation from Sparse Inputs
Realistic and smooth full-body tracking is crucial for immersive AR/VR applications. Existing systems primarily track head and hands via Head Mounted Devices (HMDs) and controllers, making the 3D full-body reconstruction in-complete. One potential approach is to generate the full-body motions from sparse inputs collected from limited sensors using a Neural Network (NN) model. In this paper, we propose a novel method based on a multi-layer perceptron (MLP) backbone that is enhanced with residual connections and a novel NN-component called Memory-Block. In particular, Memory-Block represents missing sensor data with trainable code-vectors, which are combined with the sparse signals from previous time instances to improve the temporal consistency. Furthermore, we formulate our solution as a multi-task learning problem, allowing our MLP-backbone to learn robust representations that boost accuracy. Our experiments show that our method outperforms state-of-the-art baselines by substantially reducing prediction errors. Moreover, it achieves 72 FPS on mobile HMDs that ultimately improves the accuracy-running time tradeoff.
Open AccessMem-MLP: Real-Time 3D Human Motion Generation from Sparse Inputs
Realistic and smooth full-body tracking is crucial for immersive AR/VR applications. Existing systems primarily track head and hands via Head Mounted Devices (HMDs) and controllers, making the 3D full-body reconstruction in-complete. One potential approach is to generate the full-body motions from sparse inputs collected from limited sensors using a Neural Network (NN) model. In this paper, we propose a novel method based on a multi-layer perceptron (MLP) backbone that is enhanced with residual connections and a novel NN-component called Memory-Block. In particular, Memory-Block represents missing sensor data with trainable code-vectors, which are combined with the sparse signals from previous time instances to improve the temporal consistency. Furthermore, we formulate our solution as a multi-task learning problem, allowing our MLP-backbone to learn robust representations that boost accuracy. Our experiments show that our method outperforms state-of-the-art baselines by substantially reducing prediction errors. Moreover, it achieves 72 FPS on mobile HMDs that ultimately improves the accuracy-running time tradeoff.