基於3D卷積網路於時間與設備限制下行為辨識結果探討

Study on Action Recognition Results based on 3D Convolutional Networks with Limited Time and Equipment Constraints

陳伶芳(Chen Ling-Fang)

指導教授 ： 蔡正發

國立屏東科技大學/管理學院/資訊管理系所/碩士(2018年)

https://doi.org/10.6346/THE.NPUST.MIS.020.2018.F08

摘要

行為辨識在電腦視覺中為一門重要的研究主題，在硬體設備日益進步的狀況下，使用深度學習之方法做行為辨識的研究也日漸增多，但於現實狀況下設備可能會因各種原因處處受限，且如今這步調極快的社會中，事事講求效率，如何在短時間內於有限的資源下產生較佳的結果成為最重要的事。
本篇論文選擇結構較小之C3D架構，使用遷移學習縮短其訓練時間，並以優化器、激勵函數與權重初始化方法做比較，提供使用遷移學習理論下產生之最佳組合－使用Momentum優化(使用Nesterov)搭配Xavier權重初始化與激勵ReLU，於短時間及設備限制下結果較優之模型。

關鍵字

行為辨識 ； 深度學習 ； 遷移學習 ； 卷積神經網路

並列摘要

Action recognition is an important research topic in computer vision. As hardware equipment becomes increasingly advanced, research on action recognition using deep learning methods has increased, but equipment may be affected for various reasons. Limitations and the efficiency of everything in our modern fast-paced society make how to produce better results with limited resources in a short period of time extremely important.

This thesis selects a smaller-structure C3D convolutional network, uses Transfer Learning to shorten training times, and compares the optimizer, Activation Function, and weight initialization method to provide the best combination using migration learning theory and Momentum(use Nesterov) to optimize the model with Xavier weight initialization and Activation Function RELU, which is better with short times and equipment constraints.

並列關鍵字

Action Recognition ； Deep Learning ； Transfer Learning ； Convolutional Neural Network

參考文獻

[1]S.Herath, M.Harandi, and F.Porikli, “Going deeper into action recognition: A survey,” Image and Vision Computing, vol. 60, pp. 4–21, 2017.

[2]D.Hogg, “Model-based vision: a program to see a walking person,” Image and Vision Computing, vol. 1, no. 1, pp. 5–20, 1983.

[3]A. F.Bobick and J. W.Davis, “The recognition of human movement using temporal templates,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 23, no. 3, pp. 257–267, 2001.

[4]L.Gorelick, M.Blank, E.Shechtman, M.Irani, and R.Basri, “Actions as space-time shapes,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 29, no. 12, pp. 2247–2253, 2007.

[5]P.Dollár, V.Rabaud, G.Cottrell, and S.Belongie, “Behavior recognition via sparse spatio-temporal features,” Proceedings - 2nd Joint IEEE International Workshop on Visual Surveillance and Performance Evaluation of Tracking and Surveillance, VS-PETS, vol. 2005, pp. 65–72, 2005.