Alert button

Learning Spatiotemporal Features for Infrared Action Recognition with 3D Convolutional Neural Networks

May 18, 2017
Figure 1 for Learning Spatiotemporal Features for Infrared Action Recognition with 3D Convolutional Neural Networks
Figure 2 for Learning Spatiotemporal Features for Infrared Action Recognition with 3D Convolutional Neural Networks
Figure 3 for Learning Spatiotemporal Features for Infrared Action Recognition with 3D Convolutional Neural Networks
Figure 4 for Learning Spatiotemporal Features for Infrared Action Recognition with 3D Convolutional Neural Networks

Share this with someone who'll enjoy it:

Infrared (IR) imaging has the potential to enable more robust action recognition systems compared to visible spectrum cameras due to lower sensitivity to lighting conditions and appearance variability. While the action recognition task on videos collected from visible spectrum imaging has received much attention, action recognition in IR videos is significantly less explored. Our objective is to exploit imaging data in this modality for the action recognition task. In this work, we propose a novel two-stream 3D convolutional neural network (CNN) architecture by introducing the discriminative code layer and the corresponding discriminative code loss function. The proposed network processes IR image and the IR-based optical flow field sequences. We pretrain the 3D CNN model on the visible spectrum Sports-1M action dataset and finetune it on the Infrared Action Recognition (InfAR) dataset. To our best knowledge, this is the first application of the 3D CNN to action recognition in the IR domain. We conduct an elaborate analysis of different fusion schemes (weighted average, single and double-layer neural nets) applied to different 3D CNN outputs. Experimental results demonstrate that our approach can achieve state-of-the-art average precision (AP) performances on the InfAR dataset: (1) the proposed two-stream 3D CNN achieves the best reported 77.5% AP, and (2) our 3D CNN model applied to the optical flow fields achieves the best reported single stream 75.42% AP.

Share this with someone who'll enjoy it: