Show simple item record

dc.contributor.advisorDavis, Larry S.en_US
dc.contributor.authorSANTHANAM, VENKATARAMANen_US
dc.date.accessioned2018-09-07T05:39:35Z
dc.date.available2018-09-07T05:39:35Z
dc.date.issued2018en_US
dc.identifierhttps://doi.org/10.13016/M26D5PF4K
dc.identifier.urihttp://hdl.handle.net/1903/21151
dc.description.abstractOver the past few years, deep convolutional neural network (DCNN) based approaches have been immensely successful in tackling a diverse range of object recognition problems. Popular DCNN architectures like deep residual networks (ResNets) are highly generic, not just for classification, but also for high level tasks like detection/tracking which rely on classification DCNNs as their backbone. The generality of DCNNs however doesn't extend to image-to-image(Im2Im) regression tasks (eg: super-resolution, denoising, rgb-to-depth, relighting, etc). For such tasks, DCNNs are often highly task-specific and require specific ancillary post-processing methods. The major issue plaguing the design of generic architectures for such tasks is the tradeoff between context/locality given a fixed computation/memory budget. We first present a generic DCNN architecture for Im2Im regression that can be trained end-to-end without any further machinery. Our proposed architecture, the Recursively Branched Deconvolutional Network (RBDN), which features a cheap early multi-context image representation, an efficient recursive branching scheme with extensive parameter sharing and learnable upsampling. We provide qualitative/quantitative results on 3 diverse tasks: relighting, denoising and colorization and show that our proposed RBDN architecture obtains comparable results to the state-of-the-art on each of these tasks when used off-the-shelf without any post processing or task-specific architectural modifications. Second, we focus on gradient flow and optimization in ResNets. In particular, we theoretically analyze why pre-activation(v2) ResNets outperform the original ResNets(v1) on CIFAR datasets but not on ImageNet. Our analysis reveals that although v1-ResNets lack ensembling properties, they can have a higher effective depth in comparison to v2-ResNes. Subsequently, we show that downsampling projections (while only few in number) have a significantly detrimental effect on performance. We show that by simply replacing downsampling-projections with identity-like dense-reshape shortcuts, the classification results of standard residual architectures like ResNets, ResNeXts and SE-Nets improve by up to 1.2% on ImageNet, without any increase in computational complexity (FLOPs). Finally, we present a robust non-parametric probabilistic ensemble method for multi-classification, which outperforms the state-of-the-art ensemble methods on several machine learning and computer vision datasets for object recognition with statistically significant improvements. The approach is particularly geared towards multi-classification problems with very low training data and/or a fairly high proportion of outliers, for which training end-to-end DCNNs is not very beneficial.en_US
dc.language.isoenen_US
dc.titleTowards Generalized Frameworks for Object Recognitionen_US
dc.typeDissertationen_US
dc.contributor.publisherDigital Repository at the University of Marylanden_US
dc.contributor.publisherUniversity of Maryland (College Park, Md.)en_US
dc.contributor.departmentComputer Scienceen_US
dc.subject.pqcontrolledComputer scienceen_US
dc.subject.pqcontrolledArtificial intelligenceen_US
dc.subject.pquncontrolledDeep Convolutional Neural Networksen_US
dc.subject.pquncontrolledEnsemble Methodsen_US
dc.subject.pquncontrolledGradient Flowen_US
dc.subject.pquncontrolledImage-to-Image Regressionen_US
dc.subject.pquncontrolledOptimizationen_US


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record