computer-vision-7

Human Pose Forecasting

3260 papers • 126 benchmarks • 313 datasets

Human pose forecasting is the task of detecting and predicting future human poses. ( Image credit: EgoPose )

(Image credit: Papersgraph)

Benchmarks

These leaderboards are used to track progress in human-pose-forecasting-7

Trend

Dataset

Best Model

Actions

Human3.6M

AMASS

HumanEva-I

Libraries

i

Use these libraries to find human-pose-forecasting-7 models and implementations

Datasets

Subtasks

No subtasks available.

Most implemented papers

On Human Motion Prediction Using Recurrent Neural Networks

Julieta Martinez, Michael J. Black, J. Romero•Fri May 05 2017

It is shown that, surprisingly, state of the art performance can be achieved by a simple baseline that does not attempt to model motion at all, and a simple and scalable RNN architecture is proposed that obtains state-of-the-art performance on human motion prediction.

997

Content

3DPW

HARPER

GTA-IM Dataset

Expi - common actions split

PATS

HARPER

0

Paper Graph

Learning Trajectory Dependencies for Human Motion Prediction

Hongdong Li, M. Salzmann, Wei Mao, Miaomiao Liu•Wed Aug 14 2019

A simple feed-forward deep network for motion prediction, which takes into account both temporal smoothness and spatial dependencies among human body joints, and design a new graph convolutional network to learn graph connectivity automatically.

524 0

Paper Graph

HP-GAN: Probabilistic 3D Human Motion Prediction via GAN

Zicheng Liu, E. Barsoum, J. Kender•Sun Nov 26 2017

A novel sequence-to-sequence model for probabilistic human motion prediction, trained with a modified version of improved Wasserstein generative adversarial networks (WGAN-GP), in which the model learns a probability density function of future human poses conditioned on previous poses.

362 0

Paper Graph

Deep Unsupervised Clustering with Gaussian Mixture Variational Autoencoders

Kai Arulkumaran, M. Garnelo, P. Mediano, M. J. Lee, Nat Dilokthanakul, Hugh Salimbeni, M. Shanahan•Mon Nov 07 2016

It is shown that a heuristic called minimum information constraint that has been shown to mitigate this effect in VAEs can also be applied to improve unsupervised clustering performance with this variant of the variational autoencoder model with a Gaussian mixture as a prior distribution.

701 0

Paper Graph

Structural-RNN: Deep Learning on Spatio-Temporal Graphs

Ashutosh Saxena, Ashesh Jain, Amir Zamir, S. Savarese•Mon Nov 16 2015

This paper develops a scalable method for casting an arbitrary spatio-temporal graph as a rich RNN mixture that is feedforward, fully differentiable, and jointly trainable and shows improvement over the state-of-the-art with a large margin.

1148 0

Paper Graph

Convolutional Sequence to Sequence Model for Human Dynamics

Gim Hee Lee, Zhen Zhang, Wee Sun Lee•Tue May 01 2018

This work presents a novel approach to human motion modeling based on convolutional neural networks (CNN), which is able to capture both invariant and dynamic information of human motion, which results in more accurate predictions.

361 0

Paper Graph

Accurate and Diverse Sampling of Sequences Based on a "Best of Many" Sample Objective

B. Schiele, Mario Fritz, Apratim Bhattacharyya•Thu May 31 2018

This work addresses challenges in a Gaussian Latent Variable model for sequence prediction with a "Best of Many" sample objective that leads to more accurate and more diverse predictions that better capture the true variations in real-world sequence data.

125 0

Paper Graph

The Pose Knows: Video Forecasting by Generating Pose Futures

A. Gupta, M. Hebert, Jacob Walker, Kenneth Marino•Thu Apr 27 2017

This work exploits human pose detectors as a free source of supervision and breaks the video forecasting problem into two discrete steps, and uses the structured space of pose as an intermediate representation to sidestep the problems that GANs have in generating video pixels directly.

357 0

Paper Graph

Diverse Human Motion Prediction via Gumbel-Softmax Sampling from an Auxiliary Space

Qing Zhang, Lingwei Dang, Yongwei Nie, Chengjiang Long, Guiqing Li•Thu Jul 14 2022

This paper proposes a novel sampling strategy for sampling very diverse results from an imbalanced multimodal distribution learned by a deep generative model, which incorporates a Gumbel-Softmax coefficient matrix sampling method and an aggressive diversity promoting hinge loss function.

55 0

Paper Graph

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

Dimitris N. Metaxas, Long Zhao, Xi Peng, Yu Tian, Mubbasir Kapadia•Wed Jul 25 2018

This work trains networks to learn residual motion between the current and future frames, which avoids learning motion-irrelevant details and proposes a two-stage generation framework where videos are generated from structures and then refined by temporal signals.

104 0

Paper Graph

Adding a benchmark result helps the community track progress.