Depth Pooling Based Large-scale 3D Action Recognition with Convolutional Neural Networks

Abstract

This paper proposes three simple, compact yet effective representations of depth sequences, referred to respectively as Dynamic Depth Images (DDI), Dynamic Depth Normal Images (DDNI) and Dynamic Depth Motion Normal Images (DDMNI), for both isolated and continuous action recognition. These dynamic images are constructed from a segmented sequence of depth maps using hierarchical bidirectional rank pooling to effectively capture the spatial-temporal information. Specifically, DDI exploits the dynamics of postures over time and DDNI and DDMNI exploit the 3D structural information captured by depth maps. Upon the proposed representations, a ConvNet based method is developed for action recognition. The image-based representations enable us to fine-tune the existing Convolutional Neural Network (ConvNet) models trained on image data without training a large number of parameters from scratch. The proposed method achieved the state-of-art results on three large datasets, namely, the Large-scale Continuous Gesture Recognition Dataset (means Jaccard index 0.4109), the Large-scale Isolated Gesture Recognition Dataset (59.21%), and the NTU RGB+D Dataset (87.08% cross-subject and 84.22% cross-view) even though only the depth modality was used.

arXiv commentarXiv admin note: text overlap with arXiv:1701.01814, arXiv:1608.06338

References in corpus (8)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Two-Stream Convolutional Networks for Action Recognition in Videos
• Caffe: Convolutional Architecture for Fast Feature Embedding
• Unsupervised Learning of Video Representations using LSTMs
• NTU RGB+D: A Large Scale Dataset for 3D Human Activity Analysis
• Differential Recurrent Neural Networks for Action Recognition
• Cooperative Training of Deep Aggregation Networks for RGB-D Action Recognition
• Large-scale Continuous Gesture Recognition Using Convolutional Neural Networks