“…Furthermore, Jabnoun et al, in [42], described a visual tool for VI people, based on an object recognition algorithm allowing to determine the dissimilarity between video frames; specifically, the algorithm employs the Real-Valued Local Dissimilarity Map (RVLDM) method, as a measure of frames' dissimilarity, and the Scale-Invariant Features Transform (SIFTS) keypoints extraction, for determining the objects depicted in different frames. By comparing the proposed method with similar visual substitution approaches, optimal performances have been demonstrated, in terms of computational speed in different operative conditions, such as different point-of-view, presence of occlusion, frame rotation, and different illumination.…”