Tens of thousands of years ago, humans drew sketches that we can see and identify even today. Sketches are the oldest recorded form of human communication and are still widely used. The universality of sketches supersedes that of culture and language. Despite the universal accessibility of sketches by humans, computers are unable to interpret or even correctly identify the contents of sketches drawn by humans with a practical level of accuracy.
In my thesis, I demonstrate that the accuracy of existing sketch recognition techniques can be improved by optimizing the classification criteria. Current techniques classify a 20,000 sketch crowd-sourced dataset with 56% accuracy. I classify the same dataset with 52% accuracy, but identify factors that have the greatest effect on the accuracy.
The ability for computers to identify human sketches would be useful particularly in pictionary-like games and other kinds of human-computer interaction; the concepts from sketch recognition could be extended to other kinds of object recognition.
","abstract_html":"<p>Tens of thousands of years ago, humans drew sketches that we can see and identify even today. Sketches are the oldest recorded form of human communication and are still widely used. The universality of sketches supersedes that of culture and language. Despite the universal accessibility of sketches by humans, computers are unable to interpret or even correctly identify the contents of sketches drawn by humans with a practical level of accuracy.</p> <p>In my thesis, I demonstrate that the accuracy of existing sketch recognition techniques can be improved by optimizing the classification criteria. Current techniques classify a 20,000 sketch crowd-sourced dataset with 56% accuracy. I classify the same dataset with 52% accuracy, but identify factors that have the greatest effect on the accuracy.</p> <p>The ability for computers to identify human sketches would be useful particularly in pictionary-like games and other kinds of human-computer interaction; the concepts from sketch recognition could be extended to other kinds of object recognition.</p>","abstract_has_math":false,"creators":["Steigerwald, Richard"],"institution":null,"degree_name":"MS in Computer Science","degree_level":null,"degree_discipline":"Computer Science","degree_department":null,"school":null,"contributors":["Zoë Wood"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2013,"date_issued":"2013-06-01T07:00:00Z","date_published":"2013-06-01T07:00:00Z","updated_at":"2026-07-24T01:31:17Z","subjects":["computer vision","machine learning","sketch recognition","ai","artificial intelligence","computer graphics","Graphics and Human Computer Interfaces"],"languages":[],"rights":[],"rights_urls":[],"identifier_entries":[{"key":"dc:identifier","label":"Identifier","values":["10.15368/theses.2013.50"],"render_values":[{"text":"10.15368/theses.2013.50","href":"https://doi.org/10.15368/theses.2013.50","code":true}]}]},"links":{"outbound_url":"https://digitalcommons.calpoly.edu/theses/1009","outbound_label":"Repository record","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Zoë Wood"]},{"key":"dc:creator","label":"Author","values":["Steigerwald, Richard"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.available","label":"Dc Date Available","values":["2013-07-11T07:00:00Z"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Computer Science"]},{"key":"thesis:degree_name","label":"Degree Name","values":["MS in Computer Science"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["computer vision","machine learning","sketch recognition","ai","artificial intelligence","computer graphics","Graphics and Human Computer Interfaces"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["https://digitalcommons.calpoly.edu/theses/1009","10.15368/theses.2013.50"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Tens of thousands of years ago, humans drew sketches that we can see and identify even today. Sketches are the oldest recorded form of human communication and are still widely used. The universality of sketches supersedes that of culture and language. Despite the universal accessibility of sketches by humans, computers are unable to interpret or even correctly identify the contents of sketches drawn by humans with a practical level of accuracy.
In my thesis, I demonstrate that the accuracy of existing sketch recognition techniques can be improved by optimizing the classification criteria. Current techniques classify a 20,000 sketch crowd-sourced dataset with 56% accuracy. I classify the same dataset with 52% accuracy, but identify factors that have the greatest effect on the accuracy.
The ability for computers to identify human sketches would be useful particularly in pictionary-like games and other kinds of human-computer interaction; the concepts from sketch recognition could be extended to other kinds of object recognition.
"]},{"key":"dc:title","label":"Title","values":["Computer Sketch Recognition"]}]}],"canonical_facts":{"dc:contributor":["Zoë Wood"],"dc:creator":["Steigerwald, Richard"],"dc:date.available":["2013-07-11T07:00:00Z"],"dc:description.abstract":["Tens of thousands of years ago, humans drew sketches that we can see and identify even today. Sketches are the oldest recorded form of human communication and are still widely used. The universality of sketches supersedes that of culture and language. Despite the universal accessibility of sketches by humans, computers are unable to interpret or even correctly identify the contents of sketches drawn by humans with a practical level of accuracy.
In my thesis, I demonstrate that the accuracy of existing sketch recognition techniques can be improved by optimizing the classification criteria. Current techniques classify a 20,000 sketch crowd-sourced dataset with 56% accuracy. I classify the same dataset with 52% accuracy, but identify factors that have the greatest effect on the accuracy.
The ability for computers to identify human sketches would be useful particularly in pictionary-like games and other kinds of human-computer interaction; the concepts from sketch recognition could be extended to other kinds of object recognition.
"],"dc:identifier":["https://digitalcommons.calpoly.edu/theses/1009","10.15368/theses.2013.50"],"dc:subject":["computer vision","machine learning","sketch recognition","ai","artificial intelligence","computer graphics","Graphics and Human Computer Interfaces"],"dc:title":["Computer Sketch Recognition"],"thesis:degree_discipline":["Computer Science"],"thesis:degree_name":["MS in Computer Science"]},"updated_at":"2026-07-24T01:31:17Z"}