Human Pose Forecasting

8 papers with code • 3 benchmarks • 4 datasets

Human pose forecasting is the task of detecting and predicting future human poses.

( Image credit: EgoPose )

Most implemented papers

Learning Trajectory Dependencies for Human Motion Prediction

wei-mao-2019/LearnTrajDep ICCV 2019

In this paper, we propose a simple feed-forward deep network for motion prediction, which takes into account both temporal smoothness and spatial dependencies among human body joints.

Structural-RNN: Deep Learning on Spatio-Temporal Graphs

asheshjain399/RNNexp CVPR 2016

The proposed method is generic and principled as it can be used for transforming any spatio-temporal graph through employing a certain set of well defined steps.

Learning to Forecast and Refine Residual Motion for Image-to-Video Generation

garyzhao/FRGAN ECCV 2018

We consider the problem of image-to-video translation, where an input image is translated into an output video containing motions of a single object.

Action-Agnostic Human Pose Forecasting

eddyhkchiu/pose_forecast_wacv 23 Oct 2018

In this paper, we propose a new action-agnostic method for short- and long-term human pose forecasting.

Ego-Pose Estimation and Forecasting as Real-Time PD Control

Khrylx/EgoPose ICCV 2019

We propose the use of a proportional-derivative (PD) control based policy learned via reinforcement learning (RL) to estimate and forecast 3D human pose from egocentric videos.

DLow: Diversifying Latent Flows for Diverse Human Motion Prediction

Khrylx/DLow ECCV 2020

To obtain samples from a pretrained generative model, most existing generative human motion prediction methods draw a set of independent Gaussian latent codes and convert them to motion samples.

Robust Motion In-betweening

ubisoftinc/Ubisoft-LaForge-Animation-Dataset 9 Feb 2021

To quantitatively evaluate performance on transitions and generalizations to longer time horizons, we present well-defined in-betweening benchmarks on a subset of the widely used Human3. 6M dataset and on LaFAN1, a novel high quality motion capture dataset that is more appropriate for transition generation.

Space-Time-Separable Graph Convolutional Network for Pose Forecasting

fraluca/stsgcn ICCV 2021

For the first time, STS-GCN models the human pose dynamics only with a graph convolutional network (GCN), including the temporal evolution and the spatial joint interaction within a single-graph framework, which allows the cross-talk of motion and spatial correlations.