Abstract. We propose a multi-cue based approach for recognizing human actions in still images, where relevant object regions are discovered and utilized in a weakly supervised manner. Our approach does not require any explicitly trained object detector or part/attribute annotation. Instead, a multiple instance learning approach is used over sets of object hypotheses in order to represent objects relevant to the actions. We test our method on the extensive Stanford 40 Actions dataset [1] and achieve significant performance gain compared to the state-of-the-art. Our results show that using multiple object hypotheses within multiple instance learning is effective for human action recognition in still images and such an object representation is suitable for using in conjunction with other visual features.

No abstract

A layered perceptron artificial neural network (ANN) is trained to detect positive signals corrupted with noise which, for our test, is Laplacian. Comparison of the ANN performance is made with both Neyman-Pearson optimal and linear detectors. The ANN invariably outperforms the linear detector and is shown to be nearly optimal. The optimal detector requires knowledge of signal and noise parameters. The ANN does not.

No abstract

No abstract