paper-with-me

홈 › Papers

QuaterNet: A Quaternion-based Recurrent Model for Human Motion

2018-05-16 · Dario Pavllo, David Grangier, Michael Auli

Deep learning for predicting or generating 3D human pose sequences is an active research area. Previous work regresses either joint rotations or joint positions. The former strategy is prone to error accumulation along the kinematic chain, as well as discontinuities when using Euler angle or exponential map parameterizations. The latter requires re-projection onto skeleton constraints to avoid bone stretching and invalid configurations. This work addresses both limitations. Our recurrent network, QuaterNet, represents rotations with quaternions and our loss function performs forward kinematics on a skeleton to penalize absolute position errors instead of angle errors. On short-term predictions, QuaterNet improves the state-of-the-art quantitatively. For long-term generation, our approach is qualitatively judged as realistic as recent neural strategies from the graphics literature.

📄 PDF Abstract BibTeX arXiv:1805.06485

Code (1)

facebookresearch/QuaterNet 공식 구현 pytorch

Tasks

3D Human Pose EstimationMotion EstimationPosition

Similar Papers 제목 키워드 기반

Modeling Human Motion with Quaternion-based Neural Networks

2019-01-21 · Dario Pavllo, Christoph Feichtenhofer, Michael Auli, David Grangier

Previous work on predicting or generating 3D human pose sequences regresses either joint rotations or joint positions. The former strategy is prone to error accumulation along the kinematic chain, as well as discontinuit…

PVRED: A Position-Velocity Recurrent Encoder-Decoder for Human Motion Prediction

2019-06-15 · Hongsong Wang, Jian Dong, Bin Cheng, Jiashi Feng

Human motion prediction, which aims to predict future human poses given past poses, has recently seen increased interest. Many recent approaches are based on Recurrent Neural Networks (RNN) which model human poses with e…

DecoderHuman motion predictionmotion predictionPosition+1

Quaternion recurrent neural network with real-time recurrent learning and maximum correntropy criterion

2024-02-22 · Pauline Bourigault, Dongpo Xu, Danilo P. Mandic

We develop a robust quaternion recurrent neural network (QRNN) for real-time processing of 3D and 4D data with outliers. This is achieved by combining the real-time recurrent learning (RTRL) algorithm and the maximum cor…

motion prediction

QuaMo: Quaternion Motions for Vision-based 3D Human Kinematics Capture

2026-01-27 · Cuong Le, Pavlo Melnyk, Urs Waldmann, Mårten Wadenbäck 외 arxiv

Vision-based 3D human motion capture from videos remains a challenge in computer vision. Traditional 3D pose estimation approaches often ignore the temporal consistency between frames, causing implausible and jittery mot…

3D Pose Estimation

Quaternion-Valued Recurrent Projection Neural Networks on Unit Quaternions

2020-01-30 · Marcos Eduardo Valle, Rodolfo Anibal Lobo

Hypercomplex-valued neural networks, including quaternion-valued neural networks, can treat multi-dimensional data as a single entity. In this paper, we present the quaternion-valued recurrent projection neural networks …