The last contribution of this dissertation is to develop new method to estimate the connection weights of the CNNs.
It is based on training an SVM for each kernel of the CNN.
The decision of the committee will be final, and no request for review/revision whatsoever will be entertained.
IUPRAI-ICVGIP Award Committee visit more details.
Further, in order to improve the representativeness of the extracted features, we reinforce them with a feature learning stage by means of an autoencoder model.
This latter is topped with a logistic regression layer in order to detect the presence of objects if any.Two reconstruction strategies are proposed, namely pixel-based and patch-based reconstructions.From the earlier two topics, we quantitatively demonstrate that autoencoders can play a pivotal role in terms of both (i) feature learning and (ii) reconstruction and mapping of sequential data.Each nomination should be forwarded by the thesis advisor (not the student).Nomination should be one zipped file from the thesis advisor, which should include (all information will be kept strictly confidential): Important Dates Submission deadline: 8th December 2018 Review Process The nomination letter, list of publications and summary of the thesis will be used in initial screening phase to shortlist dissertation for subsequent in-depth evaluation.Only English language versions of the thesis will be accepted.Nomination Process The deadline for nomination is 8th December, 2018.Convolutional Neural Network (CNN) is arguably the most utilized model by the computer vision community, which is reasonable thanks to its remarkable performance in object and scene recognition, with respect to traditional hand-crafted features.Nevertheless, it is evident that CNN naturally is availed in its two-dimensional version.Although it has demonstrated cutting-edge performance widely in computer vision, and particularly in object recognition and detection, deep learning is yet to find its way into other research areas.Furthermore, the performance of deep learning models has a strong dependency on the way in which these latter are designed/tailored to the problem at hand.