The first is the pre-trained mid-level representations non-fixed fine-tuning strategy.
Low-level features capture the time and frequency domain property while mid-level representations learn the composition of the action.
The large semantic gap motivates that most methods model scenes resorting to mid-level representations (e.g. objects, topics).
While most of the learning rules do not yield mirror-symmetry in the mid-level representations, we characterize a specific biologically-plausible Hebb-type learning rule that is guaranteed to generate mirror-symmetric tuning to faces tuning at intermediate levels of the architecture.
Thorough experiments conducted on standard databases show that: (1) DeepSim can accurately predict human perceived image quality and outperforms previous state-of-the-art; (2) mid-level representations are most effective for quality prediction; and (3) preprocessing, the restricted linear units and max-pooling operations are beneficial for the IQA performance.
The BoVW follows a specific pipeline: (i) extract local features from the images, (ii) learn a visual dictionary, (iii) create mid-level representations of the images using the visual dictionary, and (iv) learn a classifier using the mid-level representations.
Two networks are defined (Fig. 2): (i) a coding network U x;θ u ) parameterized by θ u that learns to extract features and cluster input instances together and (ii) a classification network D x;θ d ) parameterized by θ d that discriminates between normal and pathological mid-level representations.
In the second level, the output scores from the previous level are aggregated into the mid-level representation which is conducted to an additional ELM-E classifier.
For addressing that challenging problem, some visual and semantic attributes are usually used as mid-level representation to transfer knowledge from training classes to unseen test ones.
Building on a theory of invariance in hierarchical architectures, we propose a novel, mid-level representation for acoustical signals, using the empirical distributions of projections on a set of templates and their transformations.
In the original publication the mid-level representation used spectral resolution of five bands.
