ログイン
言語:

WEKO3

  • トップ
  • ランキング
To
lat lon distance
To

Field does not validate



インデックスリンク

インデックスツリー

メールアドレスを入力してください。

WEKO

One fine body…

WEKO

One fine body…

アイテム

{"_buckets": {"deposit": "2ae8e839-b032-446a-9343-02d62a0d0696"}, "_deposit": {"id": "2010973", "owners": [1], "pid": {"revision_id": 0, "type": "depid", "value": "2010973"}, "status": "published"}, "_oai": {"id": "oai:u-ryukyu.repo.nii.ac.jp:02010973", "sets": ["1642838338003", "1642838406845"]}, "author_link": [], "item_1617186331708": {"attribute_name": "Title", "attribute_value_mlt": [{"subitem_1551255647225": "Multi-task Learning \u3068Multi-stream \u306e\u7d71\u5408\u30e2\u30c7\u30eb\u3092\u7528\u3044\u305f\u5358\u773c\u6df1\u5ea6\u63a8\u5b9a", "subitem_1551255648112": "ja"}, {"subitem_1551255647225": "Monocular Depth Estimation using Integrated Model with Multi-task Learning and Multi-stream", "subitem_1551255648112": "en"}]}, "item_1617186419668": {"attribute_name": "Creator", "attribute_type": "creator", "attribute_value_mlt": [{"creatorNames": [{"creatorName": "\u9ad9\u5dba, \u6f6e", "creatorNameLang": "ja"}]}, {"creatorNames": [{"creatorName": "\u9060\u85e4, \u8061\u5fd7", "creatorNameLang": "ja"}]}, {"creatorNames": [{"creatorName": "Takamine, Michiru", "creatorNameLang": "en"}]}, {"creatorNames": [{"creatorName": "Endo, Satoshi", "creatorNameLang": "en"}]}]}, "item_1617186476635": {"attribute_name": "Access Rights", "attribute_value_mlt": [{"subitem_1522299639480": "open access", "subitem_1600958577026": "http://purl.org/coar/access_right/c_abf2"}]}, "item_1617186499011": {"attribute_name": "Rights", "attribute_value_mlt": [{"subitem_1522650717957": "ja", "subitem_1522651041219": "\u4eba\u5de5\u77e5\u80fd\u5b66\u4f1a2021"}]}, "item_1617186609386": {"attribute_name": "Subject", "attribute_value_mlt": [{"subitem_1522299896455": "en", "subitem_1522300014469": "Other", "subitem_1523261968819": "depth estimation"}, {"subitem_1522299896455": "en", "subitem_1522300014469": "Other", "subitem_1523261968819": "multi-task learning"}, {"subitem_1522299896455": "en", "subitem_1522300014469": "Other", "subitem_1523261968819": "multi-stream"}]}, "item_1617186626617": {"attribute_name": "Description", "attribute_value_mlt": [{"subitem_description": "Scene understanding is a central problem in a field of computer vision. Depth estimation, in particular, is one of the important applications in scene understanding, robotics, and 3-D reconstruction. Estimating a dense depth map from a single image is receiving increased attention because a monocular camera is popular, small and suitable for a wide range of environments. In addition, both multi-task learning and multi-stream, which use unlabeled information, improve the monocular depth estimation efficiently. However, there are only a few networks optimized for both of them. Therefore, in this paper, we propose a monocular depth estimation task with a multi-task and multistream network architecture. Furthermore, the integrated network which we develop makes use of depth gradient information and can be applied to both supervised and unsupervised learning. In our experiments, we confirmed that our supervised learning architecture improves the accuracy of depth estimation by 0.13 m on average. Additionally, the experimental result on unsupervised learning found that it improved structure-from-motion performance.", "subitem_description_type": "Other"}, {"subitem_description": "\u8ad6\u6587", "subitem_description_type": "Other"}]}, "item_1617186643794": {"attribute_name": "Publisher", "attribute_value_mlt": [{"subitem_1522300295150": "ja", "subitem_1522300316516": "\u4eba\u5de5\u77e5\u80fd\u5b66\u4f1a"}]}, "item_1617186702042": {"attribute_name": "Language", "attribute_value_mlt": [{"subitem_1551255818386": "jpn"}]}, "item_1617186783814": {"attribute_name": "Identifier", "attribute_value_mlt": [{"subitem_identifier_type": "HDL", "subitem_identifier_uri": "http://hdl.handle.net/20.500.12000/49790"}]}, "item_1617186920753": {"attribute_name": "Source Identifier", "attribute_value_mlt": [{"subitem_1522646500366": "ISSN", "subitem_1522646572813": "1346-8030"}, {"subitem_1522646500366": "ISSN", "subitem_1522646572813": "1346-0714"}]}, "item_1617186941041": {"attribute_name": "Source Title", "attribute_value_mlt": [{"subitem_1522650068558": "ja", "subitem_1522650091861": "\u4eba\u5de5\u77e5\u80fd\u5b66\u4f1a\u8ad6\u6587\u8a8c"}]}, "item_1617187056579": {"attribute_name": "Bibliographic Information", "attribute_value_mlt": [{"bibliographicIssueNumber": "5", "bibliographicVolumeNumber": "36"}]}, "item_1617258105262": {"attribute_name": "Resource Type", "attribute_value_mlt": [{"resourcetype": "journal article", "resourceuri": "http://purl.org/coar/resource_type/c_6501"}]}, "item_1617265215918": {"attribute_name": "Version Type", "attribute_value_mlt": [{"subitem_1522305645492": "VoR", "subitem_1600292170262": "http://purl.org/coar/version/c_970fb48d4fbd8a85"}]}, "item_1617353299429": {"attribute_name": "Relation", "attribute_value_mlt": [{"subitem_1522306287251": {"subitem_1522306382014": "DOI", "subitem_1522306436033": "https://doi.org/10.1527/tjsai.36-5_B-KC6"}}, {"subitem_1522306287251": {"subitem_1522306382014": "DOI", "subitem_1522306436033": "https://doi.org/10.1527/tjsai.36-5_B-KC6"}}]}, "item_1617605131499": {"attribute_name": "File", "attribute_type": "file", "attribute_value_mlt": [{"accessrole": "open_access", "download_preview_message": "", "file_order": 0, "filename": "36_36-5_B-KC6.pdf", "future_date_message": "", "is_thumbnail": false, "mimetype": "", "size": 0, "url": {"objectType": "fulltext", "url": "https://u-ryukyu.repo.nii.ac.jp/record/2010973/files/36_36-5_B-KC6.pdf"}, "version_id": "51e1fe5b-d31c-4253-b3bf-488c32be6d17"}]}, "item_title": "Multi-task Learning \u3068Multi-stream \u306e\u7d71\u5408\u30e2\u30c7\u30eb\u3092\u7528\u3044\u305f\u5358\u773c\u6df1\u5ea6\u63a8\u5b9a", "item_type_id": "15", "owner": "1", "path": ["1642838338003", "1642838406845"], "permalink_uri": "http://hdl.handle.net/20.500.12000/49790", "pubdate": {"attribute_name": "PubDate", "attribute_value": "2021-09-30"}, "publish_date": "2021-09-30", "publish_status": "0", "recid": "2010973", "relation": {}, "relation_version_is_last": true, "title": ["Multi-task Learning \u3068Multi-stream \u306e\u7d71\u5408\u30e2\u30c7\u30eb\u3092\u7528\u3044\u305f\u5358\u773c\u6df1\u5ea6\u63a8\u5b9a"], "weko_shared_id": -1}
  1. 学術雑誌論文
  2. その他
  1. 部局別インデックス
  2. 工学部

Multi-task Learning とMulti-stream の統合モデルを用いた単眼深度推定

http://hdl.handle.net/20.500.12000/49790
http://hdl.handle.net/20.500.12000/49790
1fd18f49-5e3e-4d8a-af9c-856b711e7437
名前 / ファイル ライセンス アクション
36_36-5_B-KC6.pdf 36_36-5_B-KC6.pdf
Item type デフォルトアイテムタイプ(フル)(1)
公開日 2021-09-30
タイトル
タイトル Multi-task Learning とMulti-stream の統合モデルを用いた単眼深度推定
言語 ja
作成者 髙嶺, 潮

× 髙嶺, 潮

ja 髙嶺, 潮

遠藤, 聡志

× 遠藤, 聡志

ja 遠藤, 聡志

Takamine, Michiru

× Takamine, Michiru

en Takamine, Michiru

Endo, Satoshi

× Endo, Satoshi

en Endo, Satoshi

アクセス権
アクセス権 open access
アクセス権URI http://purl.org/coar/access_right/c_abf2
権利情報
言語 ja
権利情報 人工知能学会2021
主題
言語 en
主題Scheme Other
主題 depth estimation
言語 en
主題Scheme Other
主題 multi-task learning
言語 en
主題Scheme Other
主題 multi-stream
内容記述
内容記述タイプ Other
内容記述 Scene understanding is a central problem in a field of computer vision. Depth estimation, in particular, is one of the important applications in scene understanding, robotics, and 3-D reconstruction. Estimating a dense depth map from a single image is receiving increased attention because a monocular camera is popular, small and suitable for a wide range of environments. In addition, both multi-task learning and multi-stream, which use unlabeled information, improve the monocular depth estimation efficiently. However, there are only a few networks optimized for both of them. Therefore, in this paper, we propose a monocular depth estimation task with a multi-task and multistream network architecture. Furthermore, the integrated network which we develop makes use of depth gradient information and can be applied to both supervised and unsupervised learning. In our experiments, we confirmed that our supervised learning architecture improves the accuracy of depth estimation by 0.13 m on average. Additionally, the experimental result on unsupervised learning found that it improved structure-from-motion performance.
内容記述タイプ Other
内容記述 論文
出版者
言語 ja
出版者 人工知能学会
言語
言語 jpn
資源タイプ
資源タイプ journal article
資源タイプ識別子 http://purl.org/coar/resource_type/c_6501
出版タイプ
出版タイプ VoR
出版タイプResource http://purl.org/coar/version/c_970fb48d4fbd8a85
識別子
識別子 http://hdl.handle.net/20.500.12000/49790
識別子タイプ HDL
関連情報
関連識別子
識別子タイプ DOI
関連識別子 https://doi.org/10.1527/tjsai.36-5_B-KC6
関連識別子
識別子タイプ DOI
関連識別子 https://doi.org/10.1527/tjsai.36-5_B-KC6
収録物識別子
収録物識別子タイプ ISSN
収録物識別子 1346-8030
収録物識別子タイプ ISSN
収録物識別子 1346-0714
収録物名
言語 ja
収録物名 人工知能学会論文誌
書誌情報
巻 36, 号 5
戻る
0
views
See details
Views

Versions

Ver.1 2022-02-01 05:34:26.596603
Show All versions

Share

Mendeley Twitter Facebook Print Addthis

Cite as

エクスポート

OAI-PMH
  • OAI-PMH JPCOAR
  • OAI-PMH DublinCore
  • OAI-PMH DDI
Other Formats
  • JSON

確認


Powered by WEKO3


Powered by WEKO3