{"id":{"repo_id":"mcmaster","oai_identifier":"oai:macsphere.mcmaster.ca:11375/27011"},"canonical_url":"https://search.ndltd.org/etd/mcmaster/oai:macsphere.mcmaster.ca:11375/27011","repository":{"repo_id":"mcmaster","name":"McMaster University","base_url":"https://macsphere.mcmaster.ca/server/oai/request"},"display":{"title":"Enabling Trimap-Free Image Matting via Multitask Learning","abstract":"Trimap-free natural image matting problem is an important computer vision task in which we extract foreground objects from given images without extra trimap input. Compared with trimap-based matting algorithms, trimap-free algorithms are easier to make false detection when the foreground object is not well defined. To solve the problem, we design a novel structure (SegMatting) to handle foreground segmentation and alpha matte prediction simultaneously, which is able to produce high-quality mattes based on RGB inputs alone. This entangled structure enables information exchange between the binary segmentation task and the alpha matte prediction task interactively, and we further design a hybrid loss to adaptively balance two tasks during the multitask learning process. Additionally, we adopt a salient object detection dataset to pretrain our network so that we could obtain a more accurate foreground segment before our training process. Experiments indicate that the proposed SegMatting qualitatively and quantitatively outperforms most previous trimap-free models with a significant margin, while remains competitive among trimap-based methods.","abstract_html":"Trimap-free natural image matting problem is an important computer vision task in which we extract foreground objects from given images without extra trimap input. Compared with trimap-based matting algorithms, trimap-free algorithms are easier to make false detection when the foreground object is not well defined. To solve the problem, we design a novel structure (SegMatting) to handle foreground segmentation and alpha matte prediction simultaneously, which is able to produce high-quality mattes based on RGB inputs alone. This entangled structure enables information exchange between the binary segmentation task and the alpha matte prediction task interactively, and we further design a hybrid loss to adaptively balance two tasks during the multitask learning process. Additionally, we adopt a salient object detection dataset to pretrain our network so that we could obtain a more accurate foreground segment before our training process. Experiments indicate that the proposed SegMatting qualitatively and quantitatively outperforms most previous trimap-free models with a significant margin, while remains competitive among trimap-based methods.","abstract_has_math":false,"creators":["LI, CHENGQI"],"institution":null,"degree_name":null,"degree_level":null,"degree_discipline":null,"degree_department":null,"school":null,"contributors":[],"advisors":["CHEN, JUN"],"committee_chairs":[],"committee_members":[],"year":2021,"date_issued":"2021","date_published":"2021","updated_at":"2026-11-18T19:33:20Z","subjects":["Natural Image Matting","Deep Learning","Trimap-Free"],"languages":["en"],"rights":[],"rights_urls":[],"identifier_entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://doi.org/10.71548/13911"],"render_values":[{"text":"https://doi.org/10.71548/13911","href":"https://doi.org/10.71548/13911","code":true}]}],"person_identifier_entries":[]},"links":{"outbound_url":"http://hdl.handle.net/11375/27011","outbound_label":"Handle","outbound_source":"dc:identifier.uri","record_url":"http://hdl.handle.net/11375/27011","record_label":"Repository record","record_kind":"handle","record_source":"dc:identifier.uri","download_url":null,"download_label":null,"download_source":null,"access_status":null,"embargo_until":null,"access_note":null},"source_record":{"url":"https://macsphere.mcmaster.ca/server/oai/request?verb=GetRecord&metadataPrefix=dim&identifier=oai%3Amacsphere.mcmaster.ca%3A11375%2F27011","prefix":"dim"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["CHEN, JUN"]},{"key":"dc:creator","label":"Author","values":["LI, CHENGQI"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2021-10-07T17:50:03Z"]},{"key":"dc:date.available","label":"Dc Date Available","values":["2021-10-07T17:50:03Z"]},{"key":"dc:date.issued","label":"Date","values":["2021"]},{"key":"dc:type","label":"Dc Type","values":["Thesis"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Natural Image Matting","Deep Learning","Trimap-Free"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language.iso","label":"Language (ISO)","values":["en"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["http://hdl.handle.net/11375/27011","https://doi.org/10.71548/13911"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Trimap-free natural image matting problem is an important computer vision task in which we extract foreground objects from given images without extra trimap input. Compared with trimap-based matting algorithms, trimap-free algorithms are easier to make false detection when the foreground object is not well defined. To solve the problem, we design a novel structure (SegMatting) to handle foreground segmentation and alpha matte prediction simultaneously, which is able to produce high-quality mattes based on RGB inputs alone. This entangled structure enables information exchange between the binary segmentation task and the alpha matte prediction task interactively, and we further design a hybrid loss to adaptively balance two tasks during the multitask learning process. Additionally, we adopt a salient object detection dataset to pretrain our network so that we could obtain a more accurate foreground segment before our training process. Experiments indicate that the proposed SegMatting qualitatively and quantitatively outperforms most previous trimap-free models with a significant margin, while remains competitive among trimap-based methods."]},{"key":"dc:description.degree","label":"Dc Description Degree","values":["Master of Science in Electrical and Computer Engineering (MSECE)"]},{"key":"dc:title","label":"Title","values":["Enabling Trimap-Free Image Matting via Multitask Learning"]},{"key":"others:access_status","label":"Others Access Status","values":["open.access"]}]}],"canonical_facts":{"dc:contributor.advisor":["CHEN, JUN"],"dc:creator":["LI, CHENGQI"],"dc:date.accessioned":["2021-10-07T17:50:03Z"],"dc:date.available":["2021-10-07T17:50:03Z"],"dc:date.issued":["2021"],"dc:description.abstract":["Trimap-free natural image matting problem is an important computer vision task in which we extract foreground objects from given images without extra trimap input. Compared with trimap-based matting algorithms, trimap-free algorithms are easier to make false detection when the foreground object is not well defined. To solve the problem, we design a novel structure (SegMatting) to handle foreground segmentation and alpha matte prediction simultaneously, which is able to produce high-quality mattes based on RGB inputs alone. This entangled structure enables information exchange between the binary segmentation task and the alpha matte prediction task interactively, and we further design a hybrid loss to adaptively balance two tasks during the multitask learning process. Additionally, we adopt a salient object detection dataset to pretrain our network so that we could obtain a more accurate foreground segment before our training process. Experiments indicate that the proposed SegMatting qualitatively and quantitatively outperforms most previous trimap-free models with a significant margin, while remains competitive among trimap-based methods."],"dc:description.degree":["Master of Science in Electrical and Computer Engineering (MSECE)"],"dc:identifier.uri":["http://hdl.handle.net/11375/27011","https://doi.org/10.71548/13911"],"dc:language.iso":["en"],"dc:subject":["Natural Image Matting","Deep Learning","Trimap-Free"],"dc:title":["Enabling Trimap-Free Image Matting via Multitask Learning"],"dc:type":["Thesis"],"others:access_status":["open.access"]},"updated_at":"2026-11-18T19:33:20Z"}