Generative model‐enhanced human motion prediction (2020-10-05T00:00:00.000000Z)

TL;DR

A new OoD benchmark is formulated, and a hybrid framework for hardening discriminative architectures to OoD failure by augmenting them with a generative model is introduced, which improves OoD robustness without sacrificing in‐distribution performance and can theoretically facilitate model interpretability.

Abstract

Abstract The task of predicting human motion is complicated by the natural heterogeneity and compositionality of actions, necessitating robustness to distributional shifts as far as out‐of‐distribution (OoD). Here, we formulate a new OoD benchmark based on the Human3.6M and Carnegie Mellon University (CMU) motion capture datasets, and introduce a hybrid framework for hardening discriminative architectures to OoD failure by augmenting them with a generative model. When applied to current state‐of‐the‐art discriminative models, we show that the proposed approach improves OoD robustness without sacrificing in‐distribution performance, and can theoretically facilitate model interpretability. We suggest human motion predictors ought to be constructed with OoD challenges in mind, and provide an extensible general framework for hardening diverse discriminative architectures to extreme distributional shift. The code is available at: https://github.com/bouracha/OoDMotion.

Authors

P. Nachev

3 papers

Ryan-Rhys Griffiths

4 papers

Anthony Bourached

2 papers

TL;DR

Abstract

Authors

References62 items

History Repeats Itself: Human Motion Prediction via Motion Attention

Dynamic Multiscale Graph Neural Networks for 3D Skeleton Based Human Motion Prediction

Your Classifier is Secretly an Energy Based Model and You Should Treat it Like One

Word-level Deep Sign Language Recognition from Video: A New Large-scale Dataset and Methods Comparison

Unsupervised Videographic Analysis of Rodent Behaviour

Structured Prediction Helps 3D Human Motion Modelling

Bayesian Variational Autoencoders for Unsupervised Out-of-Distribution Detection

Learning Trajectory Dependencies for Human Motion Prediction

Trajectory Planning and Safety Assessment of Autonomous Vehicles Based on Motion Prediction and Model Predictive Control

Adaptive Sensor Placement for Continuous Spaces

Human Motion Prediction via Learning Local Structure Representations and Temporal Dependencies

OpenPose: Realtime Multi-Person 2D Pose Estimation Using Part Affinity Fields

Deep Anomaly Detection with Outlier Exposure

Do Deep Generative Models Know What They Don't Know?

3D MRI brain tumor segmentation using autoencoder regularization

A Neural Temporal Model for Human Motion Prediction

Adversarial Geometry-Aware Human Motion Prediction

Integrating Gaze Tracking and Head-Motion Prediction for Mobile Device Authentication: A Proof of Concept

Convolutional Sequence to Sequence Model for Human Dynamics

QuaterNet: A Quaternion-based Recurrent Model for Human Motion

Automated video‐based detection of nocturnal convulsive seizures in a residential care setting

UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction

Human Motion Generative Model Using Variational Autoencoder

Long-Term On-board Prediction of People in Traffic Scenes Under Uncertainty

Automatic differentiation in PyTorch

Enhancing The Reliability of Out-of-distribution Image Detection in Neural Networks

On Human Motion Prediction Using Recurrent Neural Networks

What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision?

Deep Representation Learning for Human Motion Prediction and Classification

Variational Graph Auto-Encoders

A Baseline for Detecting Misclassified and Out-of-Distribution Examples in Neural Networks

Social LSTM: Human Trajectory Prediction in Crowded Spaces

Concrete Problems in AI Safety

A Survey of Motion Planning and Control Techniques for Self-Driving Urban Vehicles

Efficient movement representation by embedding Dynamic Movement Primitives in deep autoencoders

Structural-RNN: Deep Learning on Spatio-Temporal Graphs

Recurrent Network Models for Human Dynamics

Adam: A Method for Stochastic Optimization

Systematic review of Kinect applications in elderly care and stroke rehabilitation

Human3.6M: Large Scale Datasets and Predictive Methods for 3D Human Sensing in Natural Environments

Efficient Nonlinear Markov Models for Human Motion

Stochastic Backpropagation and Approximate Inference in Deep Generative Models

Auto-Encoding Variational Bayes

Target following with motion prediction for unmanned surface vehicle operating in cluttered environments

Anticipating human activities for reactive robotic response

Learning Spatio-Temporal Structure from RGB-D Videos for Human Activity Detection and Anticipation

Towards pervasive physical rehabilitation using Microsoft Kinect

Latent structured models for human pose estimation

Gait recognition using active shape model and motion prediction

Probabilistic Graphical Models: Principles and Techniques - Adaptive Computation and Machine Learning

The Recurrent Temporal Restricted Boltzmann Machine

Motion Prediction for Online Gaming

WiiMedia: motion analysis methods and applications using a consumer video game controller

Modeling Human Motion Using Binary Latent Variables

Respiratory motion prediction by using the adaptive neuro fuzzy inference system (ANFIS)

Practical Parameterization of Rotations Using the Exponential Map

Attention please! Econometrica

Learning Progressive Joint Propagation for Human Motion Prediction

Teaching Robots to Predict Human Motion

Game console and gaming object with motion prediction modeling and methods for use therewith

Deep Learning

All actions in blue: opacity=0

Field of Study

Journal Information

Name

Volume

Venue Information

Name

Type

URL

Alternate Names