We observed that recognition and understanding of spoken deictics were quite robust and that spoken deictics were always used in multi-modal input.
In Fig. 13, we observe that the recognition accuracy of various HMDB videos somewhat improves when we use more features to construct the codebook (codebook size remains the same at 4000).
First, we observe that the recognition time is much longer for ODN45 than for ODN25 (approximately 7 times longer).
Similarly to the previous analysis, we observe that there is mutual recognition between both groups.
One can observe that the quality of the recognition is quite good.
Comparing group differences in residuals between both tests, we observed that CPs' recognition difficulties were more pronounced in the controlled setting of the one-year recognition memory test (W = 273, p<0.001 one-sided) than in the famous face test (W = 240, p = 0.003 one-sided).
The experiment results is shown in Fig. 17, from which it can be observed that the recognition accuracy of all the other attributes reduce faster than that of the activity recognition.
Further, it is observed that the recognition rates decline significantly with increase in the pose angle of test images (e.g., the highest recognition rate of only 74.5% is noticed for FERET_A4 which contains pose variations of ±67.5° in test images).
It is observed that the recognition performances of the other personality traits are lower.
But Watson observes that a recognition of Harris' past gives one reason to regard with ambivalence judgments of blame.
