Make Skeleton-based Action Recognition Model Smaller, Faster and Better (2019-07-23T00:00:00.000000Z)

TL;DR

A Double-feature Double-motion Network (DD-Net) for skeleton-based action recognition, which can reach a super fast speed, and achieves state-of-the-art performance on experiment datasets: SHREC and JHMDB.

Abstract

Although skeleton-based action recognition has achieved great success in recent years, most of the existing methods may suffer from a large model size and slow execution speed. To alleviate this issue, we analyze skeleton sequence properties to propose a Double-feature Double-motion Network (DD-Net) for skeleton-based action recognition. By using a lightweight network structure (i.e., 0.15 million parameters), DD-Net can reach a super fast speed, as 3,500 FPS on an ordinary GPU (e.g., GTX 1080Ti), or, 2,000 FPS on an ordinary CPU (e.g., Intel E5-2620). By employing robust features, DD-Net achieves state-of-the-art performance on our experiment datasets: SHREC (i.e., hand actions) and JHMDB (i.e., body actions). Our code is on https://github.com/fandulu/DD-Net.

Authors

S. Sakti

4 papers

Satoshi Nakamura

4 papers

Fan Yang

1 papers

TL;DR

Abstract

Authors

References39 items

Simple yet efficient real-time pose-based action recognition

Real-Time 3D Hand Pose Estimation with 3D Convolutional Neural Networks

Can WiFi Estimate Person Pose?

MFA-Net: Motion Feature Augmented Network for Dynamic Hand Gesture Recognition from Skeletal Data

SlowFast Networks for Video Recognition

Spatial-Temporal Attention Res-TCN for Skeleton-Based Dynamic Hand Gesture Recognition

Convolutional Neural Networks for Multivariate Time Series Classification using both Inter- and Intra- Channel Parallel Convolutions

PoTion: Pose MoTion Representation for Action Recognition

Deep Progressive Reinforcement Learning for Skeleton-Based Action Recognition

Through-Wall Human Pose Estimation Using Radio Signals

Recognizing Human Actions as the Evolution of Pose Estimation Maps

Simple Baselines for Human Pose Estimation and Tracking

Convolutional Neural Networks and Long Short-Term Memory for skeleton-based human activity and hand gesture recognition

Fusing Geometric Features for Skeleton-Based Action Recognition Using Multilayer LSTM Networks

Action Recognition With Spatio–Temporal Visual Attention on Skeleton Image Sequences

Spatial Temporal Graph Convolutional Networks for Skeleton-Based Action Recognition

Ensemble Deep Learning for Skeleton-Based Action Recognition Using Temporal Sliding LSTM Networks

A 3 Cent Recognizer: Simple and Effective Retrieval and Classification of Mid-air Gestures from Single 3D Traces

Enhanced skeleton visualization for view invariant human action recognition

Skeleton-based action recognition using LSTM and CNN

VNect

3D Hand Gesture Recognition Using a Depth and Skeletal Dataset

Modeling Temporal Dynamics and Spatial Configurations of Actions Using Two-Stream Recurrent Neural Networks

Chained Multi-stream Networks Exploiting Pose, Motion, and Appearance for Action Classification and Detection

Joint Distance Maps Based Action Recognition With Convolutional Neural Networks

Realtime Multi-person 2D Pose Estimation Using Part Affinity Fields

An End-to-End Spatio-Temporal Attention Model for Human Action Recognition from Skeleton Data

Spatio-Temporal LSTM with Trust Gates for 3D Human Action Recognition

Skeleton-Based Dynamic Hand Gesture Recognition

Hierarchical recurrent neural network for skeleton based action recognition

Adam: A Method for Stochastic Optimization

Skeleton-augmented Human Action Understanding by Learning with Progressively Refined Data

Towards Understanding Action Recognition

Microsoft Kinect Sensor and Its Effect

Robust hand gesture recognition with kinect sensor

Learning a 3D Human Pose Distance Metric from Geometric Pose Descriptor

A Kinect-based system for physical rehabilitation: a pilot study for young adults with motor disabilities.

A Survey of Computer Vision-Based Human Motion Capture

Keras

Field of Study

Journal Information

Name

Venue Information

Name

Type

URL

Alternate Names