Fig. 8From: Action recognition using length-variable edge trajectory and spatio-temporal motion skeleton descriptorSample frames from the three unconstrained HAR datasets used in our experiments. From top to bottom: UCF Sports, YouTube, and HMDB51Back to article page