<?xml version="1.0" encoding="UTF-8"?><?xml-stylesheet type="text/xsl" href="static/style.xsl"?><OAI-PMH xmlns="http://www.openarchives.org/OAI/2.0/" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.openarchives.org/OAI/2.0/ http://www.openarchives.org/OAI/2.0/OAI-PMH.xsd"><responseDate>2026-11-22T18:57:11Z</responseDate><request verb="GetRecord" identifier="oai:uwspace.uwaterloo.ca:10012/17430" metadataPrefix="dim">https://dspacemainprd01.lib.uwaterloo.ca/server/oai/request</request><GetRecord><record><header><identifier>oai:uwspace.uwaterloo.ca:10012/17430</identifier><datestamp>2025-03-26T23:34:47Z</datestamp><setSpec>com_10012_9924</setSpec><setSpec>com_10012_1</setSpec><setSpec>col_10012_9930</setSpec><setSpec>col_10012_6</setSpec></header><metadata><dim:dim xmlns:dim="http://www.dspace.org/xmlns/dspace/dim" xmlns:doc="http://www.lyncode.com/xoai" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://www.dspace.org/xmlns/dspace/dim http://www.dspace.org/schema/dim.xsd">
   <dim:field mdschema="dc" element="contributor" qualifier="advisor">Veksler, Olga</dim:field>
   <dim:field mdschema="dc" element="contributor" qualifier="author" authority="fe0616b4-2524-438f-91dc-e65f9d2ada03" confidence="600">Ji, Zongliang</dim:field>
   <dim:field mdschema="dc" element="date" qualifier="accessioned">2021-09-21T00:02:07Z</dim:field>
   <dim:field mdschema="dc" element="date" qualifier="available">2021-09-21T00:02:07Z</dim:field>
   <dim:field mdschema="dc" element="date" qualifier="issued">2021-09-20</dim:field>
   <dim:field mdschema="dc" element="date" qualifier="submitted">2021-09-13</dim:field>
   <dim:field mdschema="dc" element="identifier" qualifier="uri">http://hdl.handle.net/10012/17430</dim:field>
   <dim:field mdschema="dc" element="description" qualifier="abstract" lang="en">Weakly supervised segmentation signi cantly reduces user annotation e ort. Recently,&#xd;
regularized loss was proposed for single object class segmentation under image-level weak&#xd;
supervision. Regularized loss consists of several components. Each component, if used in&#xd;
isolation, would lead to some trivial solution. However, a weighted combination of the loss&#xd;
components introduces a balance between the individual biases. The weight of each component&#xd;
in regularized loss is controlled by a hyperparameter. We propose an approach that&#xd;
searches for regularized loss hyperparameters. The main idea is to set the most important&#xd;
regularized loss component to a high weight while ensuring the other loss components are&#xd;
set to weights just su ciently high to prevent the trivial solution favoured by the most&#xd;
important component. Our approach results in a signi cantly improved performance over&#xd;
prior work with  xed hyperparameters and improves the state of the art in salient and&#xd;
semantic image level supervised segmentation.&#xd;
In addition to image level weak supervision, we propose a new approach for semantic&#xd;
segmentation with weak supervision using bounding box annotations. Our new approach to&#xd;
weak supervision from bounding boxes also makes use of hyperparameter search regularized&#xd;
loss. Previous work on weak supervision from bounding boxes constructs pseudo-ground&#xd;
truth by segmenting each box into the object and the background for each box independently&#xd;
from all the other boxes in the dataset. We argue that the collection of boxes for&#xd;
the same class naturally provides a dataset from which we can learn the appearance of that&#xd;
object class. Learning a good appearance model, in turn, leads to a better segmentation of&#xd;
each individual box. Thus for each class, we propose to train a segmentation CNN as from&#xd;
the dataset consisting of the bounding boxes for that class using our proposed single object&#xd;
approach. After we train these single-class CNNs, we apply them back to the training&#xd;
bounding boxes to obtain object/background segmentations and merge them to construct&#xd;
pseudo-ground truth. The obtained pseudo-ground truth is used for training a standard&#xd;
segmentation CNN. We improve the state of the art on Pascal VOC 2012 benchmark in&#xd;
bounding box weak supervision setting.</dim:field>
   <dim:field mdschema="dc" element="language" qualifier="iso" lang="en">en</dim:field>
   <dim:field mdschema="dc" element="publisher" lang="en">University of Waterloo</dim:field>
   <dim:field mdschema="dc" element="subject" lang="en">Computer Vision</dim:field>
   <dim:field mdschema="dc" element="subject" lang="en">Artificial Intelligence</dim:field>
   <dim:field mdschema="dc" element="subject" lang="en">Machine Learning</dim:field>
   <dim:field mdschema="dc" element="subject" lang="en">Semantic Segmentation</dim:field>
   <dim:field mdschema="dc" element="title" lang="en">Weakly-supervised Semantic Segmentation with Regularized Loss Hyperparameter Search</dim:field>
   <dim:field mdschema="dc" element="type" lang="en">Master Thesis</dim:field>
   <dim:field mdschema="dc" element="pending">false</dim:field>
   <dim:field mdschema="uws-etd" element="degree" qualifier="department" lang="en">David R. Cheriton School of Computer Science</dim:field>
   <dim:field mdschema="uws-etd" element="degree" qualifier="discipline" lang="en">Computer Science</dim:field>
   <dim:field mdschema="uws-etd" element="degree" qualifier="grantor" lang="en">University of Waterloo</dim:field>
   <dim:field mdschema="uws-etd" element="degree" lang="en">Master of Mathematics</dim:field>
   <dim:field mdschema="uws-etd" element="embargo" qualifier="terms" lang="en">0</dim:field>
   <dim:field mdschema="uws" element="contributor" qualifier="advisor">Veksler, Olga</dim:field>
   <dim:field mdschema="uws" element="contributor" qualifier="affiliation1" lang="en">Faculty of Mathematics</dim:field>
   <dim:field mdschema="uws" element="published" qualifier="city" lang="en">Waterloo</dim:field>
   <dim:field mdschema="uws" element="published" qualifier="country" lang="en">Canada</dim:field>
   <dim:field mdschema="uws" element="published" qualifier="province" lang="en">Ontario</dim:field>
   <dim:field mdschema="uws" element="typeOfResource" lang="en">Text</dim:field>
   <dim:field mdschema="uws" element="peerReviewStatus" lang="en">Unreviewed</dim:field>
   <dim:field mdschema="uws" element="scholarLevel" lang="en">Graduate</dim:field>
   <dim:field mdschema="others" element="access-status">open.access</dim:field>
</dim:dim>
</metadata></record></GetRecord></OAI-PMH>