Results - Details
Search command : Author="Στεφανίδης"
And Author="Κωνσταντίνος"
Current Record: 16 of 75
|
Identifier |
000446243 |
Title |
Semantically annotated cooking procedures for an intelligent kitchen environment |
Alternative Title |
Σημασιολογικά επισημειωμένες μαγειρικές διαδικασίες για το περιβάλλον μιας ευφυούς κουζίνας |
Author
|
Κονδυλάκης, Γιώργος Β.
|
Thesis advisor
|
Στεφανίδης, Κωνσταντίνος
|
Reviewer
|
Ζαμπούλης, Ξενοφών
Παρταράκης, Ν.
|
Abstract |
Food preparation is one of the most essential tasks in daily life and it involves a large number
of physical interactions between hands and utensils, ingredients, etc. The fundamental unit in
the food preparation activity is the concept of the recipe. The recipe describes the cooking
process, the way how to make a dish as a sequential order of cooking steps. Frequently,
following these steps can be an extremely complicated process, which requires coordination,
monitoring and execution of multiple tasks simultaneously. The complexity of such a task can
be magnified, if one thinks that an individual can slightly modify a common recipe, thus
contributing to the general multiplicity of the enterprise. With the huge advances of
technology and the increasing interest on assisting humans in their everyday lives, various
attempts have been made to enhance the cooking process by integrating multimedia and
embedding technologies in various stages. Computer vision can be used in order to establish
spatial relationships and enable reasoning about the interaction between visual entities.
This Thesis introduces a cooking assistance system that provides the user with guidance in the
accomplishment of a recipe, while validating the correct execution of each step. All the
necessary information regarding the recipe and the spatial arrangement of ingredients and
utensils is imported in the form of a single configuration file, which is construed and analyzed
by the system. Specifically, the system designed in the context of this work can provide the
user with guidance in carrying out a recipe through the appropriate messages that appear in
a panel, as a form of an interface with the user. It can also validate the overall correctness of
the cooking procedure by utilizing the following information: (a) the correct sequence of steps
which comprise a specific recipe, obtainable from the configuration file; (b) the detection of
the ingredients used and the corresponding utensils; (c) the overall motion estimation of the
scene; (d) the calculation of relative positions of objects involved in the cooking procedure,
with respect to other objects.
In order to explore the benefits and limitations of the developed system, two evaluations have
been conducted. The first experiment included three motion detection algorithms. The best
algorithm achieved 100% precision even for small lighting variations. The second experiment
included detection of objects in unseen images during a real recipe preparation. The algorithm
was able to detect almost all of the objects and at all video frames correctly. An exception was
observed when the algorithm was presented with transparent objects (class mismatch) and
objects that appeared inside other objects (false positives).
|
Language |
English |
Subject |
Ambient environment |
|
Computer vision |
|
HCI |
|
Machine learning |
|
Αλληλεπίδραση ανθρώπου-υπολογιστή |
|
Μηχανική μάθηση |
|
Υπολογιστική όραση |
Issue date |
2022-03-18 |
Collection
|
School/Department--School of Sciences and Engineering--Department of Computer Science--Post-graduate theses
|
|
Type of Work--Post-graduate theses
|
Permanent Link |
https://elocus.lib.uoc.gr//dlib/2/4/c/metadata-dlib-1646129640-442707-23870.tkl
|
Views |
461 |