Multi-view feature engineering and learning

Domain-size pooling in local descriptors: DSP-SIFT

“…A single, un-normalized cell of the "scale-invariant feature transform" SIFT [27] and its variants [2,8,10] can be written compactly as a formula [12,46]:…”

“…Exceptions include [4], where intrinsic and nuisance variability are combined and abstracted into the variance and distance between the means of scalar random variables in a binary classification task. For more general settings, the goals of reducing nuisance variability while preserving intrinsic variability is elusive as a single image does not afford the ability to separate the two [12].…”

“…However, we point out that proper marginalization requires the availability of multiple images of the same scene, and therefore cannot be performed in a single image. While most local image descriptors are computed from a single image, exceptions include [12,25]. Of course, multiple images can be "hallucinated" from one, but the resulting pooling operation can only achieve invariance to modeled transformations.…”

“…These are very crude approximations but nevertheless implicit to most local descriptors. In particular, camera rotations are not addressed in this work, although others have done so [12]. .…”

Feature encoding in band-limited distributed surveillance systems

“…Moreover, some variant works such as [14] focused on multi-view feature engineering and learning (e.g., introducing some new descriptors such as Multi-View HOG). However, such solutions are known to break down easily when the camera transformation is large or when the features are extracted from low-quality images.…”