Journal article Open Access

First-Person Activity Recognition from Micro-Action Representations using Convolutional Neural Networks and Object Flow Histograms

Panagiotis Giannakeris; Panagiotis C. Petrantonakis; Konstantinos Avgerinakis; Stefanos Vrochidis; Ioannis Kompatsiaris

A novel frst-person human activity recognition framework is pro-posed in this work. Our proposed methodology is inspired by the central role moving objects have in egocentric activity videos. Using a Deep Convolutional
Neural Network we detect objects and develop discriminant object flow his-tograms in order to represent fine-grained micro-actions during short temporal windows. Our framework is based on the assumption that large scale activities
are synthesized by fine-grained micro-actions. We gather all the micro-actions and perform Gaussian Mixture Model clusterization, so as to build a micro-action vocabulary that is later used in a Fisher encoding schema. Results
show that our method can reach 60% recognition rate on the benchmark ADL dataset. The capabilities of the proposed framework are also showcased by profoundly evaluating for a great deal of hyper-parameters and comparing to
other State-of-the-Art works.

Files (3.2 MB)
Name Size
3.2 MB Download
All versions This version
Views 7272
Downloads 167167
Data volume 533.7 MB533.7 MB
Unique views 6565
Unique downloads 162162


Cite as