PSHead: 3D Head Reconstruction from a Single Image with Diffusion Prior and Self‐Enhancement
Text‐to‐3D avatar generation has shown that diffusion models trained on general objects can capture head structure. However, image‐to‐3D avatar that creates a high‐fidelity 3D avatar from a single image remains challenging due to additional constraints. It requires recovering a detailed 3D represent...
Uloženo v:
| Vydáno v: | Computer graphics forum |
|---|---|
| Hlavní autoři: | , , , , |
| Médium: | Journal Article |
| Jazyk: | angličtina |
| Vydáno: |
01.10.2025
|
| ISSN: | 0167-7055, 1467-8659 |
| On-line přístup: | Získat plný text |
| Tagy: |
Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
|
| Abstract | Text‐to‐3D avatar generation has shown that diffusion models trained on general objects can capture head structure. However, image‐to‐3D avatar that creates a high‐fidelity 3D avatar from a single image remains challenging due to additional constraints. It requires recovering a detailed 3D representation from limited cues while capturing complex facial features like wrinkles and hair. To address these challenges, we introduce PSHead, a coarse‐to‐fine framework guided by both object and face priors, to produce a Gaussian‐based 3D avatar for a single frontal‐view reference image. In the coarse stage, we create an initial 3D representation by applying diffusion models trained for general object generation, using Score Distillation Sampling losses over novel views. This approach marks the first integration of text‐to‐image, image‐to‐image, and text‐to‐video diffusion priors, with insights into each module's contribution to learning a 3D representation. In the fine stage, we refine this representation with pretrained face generation models, which denoise rendered images and use these refined outputs as supervision to further improve 3D detail fidelity. Leveraging the versatility of 2D objects prior, PSHead is robust across various different face framings. Our method outperforms existing approaches on in‐the‐wild images, proving its robustness and ability to capture intricate details without the need for extensive 3D supervision. |
|---|---|
| AbstractList | Text‐to‐3D avatar generation has shown that diffusion models trained on general objects can capture head structure. However, image‐to‐3D avatar that creates a high‐fidelity 3D avatar from a single image remains challenging due to additional constraints. It requires recovering a detailed 3D representation from limited cues while capturing complex facial features like wrinkles and hair. To address these challenges, we introduce PSHead, a coarse‐to‐fine framework guided by both object and face priors, to produce a Gaussian‐based 3D avatar for a single frontal‐view reference image. In the coarse stage, we create an initial 3D representation by applying diffusion models trained for general object generation, using Score Distillation Sampling losses over novel views. This approach marks the first integration of text‐to‐image, image‐to‐image, and text‐to‐video diffusion priors, with insights into each module's contribution to learning a 3D representation. In the fine stage, we refine this representation with pretrained face generation models, which denoise rendered images and use these refined outputs as supervision to further improve 3D detail fidelity. Leveraging the versatility of 2D objects prior, PSHead is robust across various different face framings. Our method outperforms existing approaches on in‐the‐wild images, proving its robustness and ability to capture intricate details without the need for extensive 3D supervision. |
| Author | Yang, Jing Zhong, Fangcheng Wu, Tianhan Fogarty, Kyle Oztireli, Cengiz |
| Author_xml | – sequence: 1 givenname: Jing orcidid: 0000-0002-8794-4842 surname: Yang fullname: Yang, Jing organization: Department of Computer Science and Technology University of Cambridge Cambridge UK – sequence: 2 givenname: Tianhan orcidid: 0000-0002-3807-5839 surname: Wu fullname: Wu, Tianhan organization: Department of Computer Science and Technology University of Cambridge Cambridge UK – sequence: 3 givenname: Kyle orcidid: 0000-0002-1888-4006 surname: Fogarty fullname: Fogarty, Kyle organization: Department of Computer Science and Technology University of Cambridge Cambridge UK – sequence: 4 givenname: Fangcheng orcidid: 0000-0001-5964-5282 surname: Zhong fullname: Zhong, Fangcheng organization: Department of Computer Science and Technology University of Cambridge Cambridge UK – sequence: 5 givenname: Cengiz orcidid: 0000-0002-4700-2236 surname: Oztireli fullname: Oztireli, Cengiz organization: Department of Computer Science and Technology University of Cambridge Cambridge UK |
| BookMark | eNotkL9OwzAYxC1UJNLCwBt4ZUj5nNhJzIb6h1aqREVgJXLtz21Q4iA7FWLjEXhGnoSWcsPdDacbfkMycJ1DQq4ZjNlBt3prxzkkuTwjEeNZHheZkAMSATv0HIS4IMMQ3gCA55mIyOu6XKAydzSd0mOhT6g7F3q_133dOWp911JFy9ptG6TLVm2RftT9jk5ra_fhOFn7uvNUOUNLbOzP1_fM7ZTT2KLrL8m5VU3Aq_8ckZf57HmyiFePD8vJ_SrWSZL1sS1QciOSDaJQzECSKuAZF9pyDrlm2mwkbmQhCwHSSjCJLLhlaFV68hG5Of1q34Xg0Vbvvm6V_6wYVEcw1QFM9Qcm_QVeI1kN |
| Cites_doi | 10.1109/CVPR.2019.00482 10.1109/CVPR52688.2022.01042 10.1109/CVPR52729.2023.00816 10.1145/3641519.3657512 10.1109/CVPR52688.2022.01565 10.1007/978-3-031-73226-3_19 10.1109/CVPR.2017.19 10.1109/3DV66043.2025.00092 10.1109/CVPR52733.2024.00097 10.1109/CVPR52733.2024.02022 10.1109/CVPR52729.2023.00041 10.1109/CVPR42600.2020.00525 10.1109/CVPR52733.2024.00437 10.1145/3592433 10.1109/ICCV51070.2023.02086 10.1109/CVPR52733.2024.00058 10.1109/CVPR52733.2024.00647 10.1109/WACV45572.2020.9093363 10.1145/3610548.3618153 10.1109/CVPR52729.2023.02155 10.1109/ICCV51070.2023.00853 10.1145/3641519.3657499 10.1109/3DV62453.2024.00150 10.1109/CVPR52729.2023.02007 10.1109/3DV62453.2024.00152 10.1007/978-3-030-01264-9_47 10.1145/3588432.3591503 10.1109/ICCV51070.2023.02033 10.1145/3658162 10.1109/CVPR52729.2023.00443 10.1109/CVPR52733.2024.00635 10.1007/978-3-031-72992-8_15 10.1109/CVPR.2018.00068 10.1109/CVPR52729.2023.01211 10.1007/978-3-319-46454-1_34 10.1109/CVPR52729.2023.02017 10.1145/3592460 10.1109/ICCV.2015.425 10.1109/3DV62453.2024.00151 10.1609/aaai.v39i3.32350 10.1145/3641519.3657428 10.1109/CVPR52733.2024.00986 10.1145/3687937 10.1007/978-3-030-58452-8_24 10.1007/978-3-031-73411-3_9 |
| ContentType | Journal Article |
| DBID | AAYXX CITATION |
| DOI | 10.1111/cgf.70279 |
| DatabaseName | CrossRef |
| DatabaseTitle | CrossRef |
| DatabaseTitleList | CrossRef |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Engineering |
| EISSN | 1467-8659 |
| ExternalDocumentID | 10_1111_cgf_70279 |
| GroupedDBID | .3N .4S .DC .GA .Y3 05W 0R~ 10A 15B 1OB 1OC 29F 31~ 33P 3SF 4.4 50Y 50Z 51W 51X 52M 52N 52O 52P 52S 52T 52U 52W 52X 5GY 5HH 5LA 5VS 66C 6J9 702 7PT 8-0 8-1 8-3 8-4 8-5 8UM 8VB 930 A03 AAESR AAEVG AAHQN AAMMB AAMNL AANHP AANLZ AAONW AASGY AAXRX AAYCA AAYXX AAZKR ABCQN ABCUV ABDBF ABDPE ABEML ABPVW ACAHQ ACBWZ ACCZN ACFBH ACGFS ACPOU ACRPL ACSCC ACUHS ACXBN ACXQS ACYXJ ADBBV ADEOM ADIZJ ADKYN ADMGS ADMLS ADNMO ADOZA ADXAS ADZMN AEFGJ AEGXH AEIGN AEIMD AEMOZ AENEX AEUYR AEYWJ AFBPY AFEBI AFFNX AFFPM AFGKR AFWVQ AFZJQ AGHNM AGQPQ AGXDD AGYGG AHBTC AHEFC AHQJS AIDQK AIDYY AIQQE AITYG AIURR AJXKR AKVCP ALAGY ALMA_UNASSIGNED_HOLDINGS ALVPJ AMBMR AMYDB ARCSS ASPBG ATUGU AUFTA AVWKF AZBYB AZFZN AZVAB BAFTC BDRZF BFHJK BHBCM BMNLL BMXJE BNHUX BROTX BRXPI BY8 CAG CITATION COF CS3 CWDTD D-E D-F DCZOG DPXWK DR2 DRFUL DRSTM DU5 EAD EAP EBA EBO EBR EBS EBU EDO EJD EMK EST ESX F00 F01 F04 F5P FEDTE FZ0 G-S G.N GODZA H.T H.X HF~ HGLYW HVGLF HZI HZ~ I-F IHE IX1 J0M K1G K48 LATKE LC2 LC3 LEEKS LH4 LITHE LOXES LP6 LP7 LUTES LW6 LYRES MEWTI MK4 MRFUL MRSTM MSFUL MSSTM MXFUL MXSTM N04 N05 N9A NF~ O66 O8X O9- OIG P2W P2X P4D PALCI PQQKQ Q.N Q11 QB0 QWB R.K RDJ RIWAO RJQFR ROL RX1 SAMSI SUPJJ TH9 TN5 TUS UB1 V8K W8V W99 WBKPD WIH WIK WOHZO WQJ WXSBR WYISQ WZISG XG1 ZL0 ZZTAW ~IA ~IF ~WT |
| ID | FETCH-LOGICAL-c226t-f8e94d52bee5a1d023a04645cf4407c1cdb9eb9898509f90d2984f1efa3f1efa3 |
| ISICitedReferencesCount | 0 |
| ISICitedReferencesURI | http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001584839800001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| ISSN | 0167-7055 |
| IngestDate | Sat Nov 29 07:22:16 EST 2025 |
| IsDoiOpenAccess | false |
| IsOpenAccess | true |
| IsPeerReviewed | true |
| IsScholarly | true |
| Language | English |
| LinkModel | OpenURL |
| MergedId | FETCHMERGED-LOGICAL-c226t-f8e94d52bee5a1d023a04645cf4407c1cdb9eb9898509f90d2984f1efa3f1efa3 |
| ORCID | 0000-0001-5964-5282 0000-0002-3807-5839 0000-0002-4700-2236 0000-0002-8794-4842 0000-0002-1888-4006 |
| OpenAccessLink | https://doi.org/10.1111/cgf.70279 |
| ParticipantIDs | crossref_primary_10_1111_cgf_70279 |
| PublicationCentury | 2000 |
| PublicationDate | 2025-10-01 |
| PublicationDateYYYYMMDD | 2025-10-01 |
| PublicationDate_xml | – month: 10 year: 2025 text: 2025-10-01 day: 01 |
| PublicationDecade | 2020 |
| PublicationTitle | Computer graphics forum |
| PublicationYear | 2025 |
| References | e_1_2_10_23_2 e_1_2_10_44_2 e_1_2_10_21_2 e_1_2_10_42_2 e_1_2_10_40_2 Bhattarai A. R. (e_1_2_10_5_2) 2024 e_1_2_10_2_2 e_1_2_10_18_2 e_1_2_10_39_2 e_1_2_10_53_2 e_1_2_10_4_2 e_1_2_10_16_2 e_1_2_10_37_2 e_1_2_10_6_2 e_1_2_10_14_2 Li P. (e_1_2_10_29_2) 2024; 37 e_1_2_10_35_2 e_1_2_10_11_2 e_1_2_10_34_2 e_1_2_10_8_2 Kolotouros N. (e_1_2_10_24_2) 2024 e_1_2_10_32_2 e_1_2_10_30_2 Yariv L. (e_1_2_10_58_2) 2020; 33 Yu Z. (e_1_2_10_59_2) 2024; 43 Saharia C. (e_1_2_10_49_2) 2022 e_1_2_10_61_2 e_1_2_10_63_2 e_1_2_10_48_2 e_1_2_10_65_2 e_1_2_10_25_2 e_1_2_10_46_2 e_1_2_10_67_2 e_1_2_10_22_2 e_1_2_10_45_2 e_1_2_10_68_2 e_1_2_10_20_2 e_1_2_10_43_2 e_1_2_10_41_2 Han X. (e_1_2_10_17_2) 2024 Trevithick A. (e_1_2_10_51_2) 2023 e_1_2_10_19_2 e_1_2_10_3_2 e_1_2_10_52_2 e_1_2_10_15_2 e_1_2_10_54_2 e_1_2_10_7_2 e_1_2_10_13_2 e_1_2_10_36_2 e_1_2_10_56_2 e_1_2_10_9_2 e_1_2_10_57_2 e_1_2_10_12_2 e_1_2_10_33_2 e_1_2_10_10_2 e_1_2_10_31_2 e_1_2_10_50_2 Müller T. (e_1_2_10_38_2) 2022 Wu Y. (e_1_2_10_55_2) 2024; 43 Li T. (e_1_2_10_27_2) 2017; 36 e_1_2_10_60_2 e_1_2_10_28_2 e_1_2_10_62_2 e_1_2_10_26_2 e_1_2_10_64_2 e_1_2_10_47_2 e_1_2_10_66_2 |
| References_xml | – ident: e_1_2_10_15_2 doi: 10.1109/CVPR.2019.00482 – ident: e_1_2_10_44_2 doi: 10.1109/CVPR52688.2022.01042 – ident: e_1_2_10_39_2 doi: 10.1109/CVPR52729.2023.00816 – year: 2024 ident: e_1_2_10_24_2 article-title: Dreamhuman: Animatable 3 D avatars from text publication-title: Advances in Neural Information Processing Systems – ident: e_1_2_10_9_2 – ident: e_1_2_10_35_2 doi: 10.1145/3641519.3657512 – ident: e_1_2_10_8_2 doi: 10.1109/CVPR52688.2022.01565 – year: 2024 ident: e_1_2_10_17_2 article-title: Headsculpt: Crafting 3D head avatars with text publication-title: Advances in Neural Information Processing Systems – ident: e_1_2_10_28_2 doi: 10.1007/978-3-031-73226-3_19 – ident: e_1_2_10_11_2 – year: 2022 ident: e_1_2_10_38_2 article-title: Instant neural graphics primitives with a multiresolution hash encoding publication-title: ACM Transactions on Graphics – ident: e_1_2_10_32_2 doi: 10.1109/CVPR.2017.19 – ident: e_1_2_10_42_2 – ident: e_1_2_10_66_2 doi: 10.1109/3DV66043.2025.00092 – ident: e_1_2_10_6_2 doi: 10.1109/CVPR52733.2024.00097 – ident: e_1_2_10_13_2 doi: 10.1109/CVPR52733.2024.02022 – ident: e_1_2_10_50_2 – volume: 33 start-page: 2492 year: 2020 ident: e_1_2_10_58_2 article-title: Multiview neural surface reconstruction by disentangling geometry and appearance publication-title: Advances in Neural Information Processing Systems – ident: e_1_2_10_60_2 doi: 10.1109/CVPR52729.2023.00041 – ident: e_1_2_10_14_2 doi: 10.1109/CVPR42600.2020.00525 – ident: e_1_2_10_19_2 doi: 10.1109/CVPR52733.2024.00437 – ident: e_1_2_10_26_2 doi: 10.1145/3592433 – ident: e_1_2_10_53_2 doi: 10.1109/ICCV51070.2023.02086 – ident: e_1_2_10_56_2 – year: 2022 ident: e_1_2_10_49_2 article-title: Photorealistic text‐to‐image diffusion models with deep language understanding publication-title: Advances in Neural Information Processing Systems – ident: e_1_2_10_18_2 doi: 10.1109/CVPR52733.2024.00058 – ident: e_1_2_10_25_2 doi: 10.1109/CVPR52733.2024.00647 – ident: e_1_2_10_33_2 – ident: e_1_2_10_52_2 – volume: 37 start-page: 55975 year: 2024 ident: e_1_2_10_29_2 article-title: Era3d: High‐resolution multiview diffusion using efficient row‐wise attention publication-title: Advances in Neural Information Processing Systems – ident: e_1_2_10_48_2 doi: 10.1109/WACV45572.2020.9093363 – ident: e_1_2_10_2_2 doi: 10.1145/3610548.3618153 – ident: e_1_2_10_46_2 doi: 10.1109/CVPR52729.2023.02155 – ident: e_1_2_10_34_2 doi: 10.1109/ICCV51070.2023.00853 – ident: e_1_2_10_12_2 doi: 10.1145/3641519.3657499 – ident: e_1_2_10_36_2 doi: 10.1109/3DV62453.2024.00150 – ident: e_1_2_10_3_2 doi: 10.1109/CVPR52729.2023.02007 – ident: e_1_2_10_22_2 doi: 10.1109/3DV62453.2024.00152 – ident: e_1_2_10_41_2 doi: 10.1007/978-3-030-01264-9_47 – ident: e_1_2_10_47_2 doi: 10.1145/3588432.3591503 – ident: e_1_2_10_54_2 – ident: e_1_2_10_7_2 doi: 10.1109/ICCV51070.2023.02033 – volume: 43 start-page: 1 issue: 4 year: 2024 ident: e_1_2_10_55_2 article-title: Portrait3D: Text‐guided high‐quality 3D portrait generation using pyramid representation and GANs prior publication-title: ACM Transactions on Graphics (TOG) doi: 10.1145/3658162 – ident: e_1_2_10_57_2 doi: 10.1109/CVPR52729.2023.00443 – ident: e_1_2_10_31_2 – ident: e_1_2_10_37_2 doi: 10.1109/CVPR52733.2024.00635 – ident: e_1_2_10_23_2 doi: 10.1007/978-3-031-72992-8_15 – ident: e_1_2_10_68_2 – ident: e_1_2_10_63_2 doi: 10.1109/CVPR.2018.00068 – ident: e_1_2_10_65_2 doi: 10.1109/CVPR52729.2023.01211 – volume: 36 start-page: 194 issue: 6 year: 2017 ident: e_1_2_10_27_2 article-title: Learning a model of facial shape and expression from 4D scans publication-title: ACM Transactions on Graphics – ident: e_1_2_10_61_2 – ident: e_1_2_10_45_2 – ident: e_1_2_10_4_2 doi: 10.1007/978-3-319-46454-1_34 – ident: e_1_2_10_67_2 doi: 10.1109/CVPR52729.2023.02017 – year: 2023 ident: e_1_2_10_51_2 article-title: Real‐time radiance fields for single‐image portrait view synthesis publication-title: ACM Transactions on Graphics doi: 10.1145/3592460 – ident: e_1_2_10_30_2 doi: 10.1109/ICCV.2015.425 – ident: e_1_2_10_62_2 doi: 10.1109/3DV62453.2024.00151 – ident: e_1_2_10_20_2 doi: 10.1609/aaai.v39i3.32350 – volume-title: WACV year: 2024 ident: e_1_2_10_5_2 – ident: e_1_2_10_21_2 doi: 10.1145/3641519.3657428 – ident: e_1_2_10_10_2 doi: 10.1109/CVPR52733.2024.00986 – volume: 43 start-page: 1 issue: 6 year: 2024 ident: e_1_2_10_59_2 article-title: Gaussian opacity fields: Efficient adaptive surface reconstruction in unbounded scenes publication-title: ACM Transactions on Graphics (ToG) doi: 10.1145/3687937 – ident: e_1_2_10_43_2 – ident: e_1_2_10_16_2 – ident: e_1_2_10_40_2 doi: 10.1007/978-3-030-58452-8_24 – ident: e_1_2_10_64_2 doi: 10.1007/978-3-031-73411-3_9 |
| SSID | ssj0004765 |
| Score | 2.446709 |
| SecondaryResourceType | online_first |
| Snippet | Text‐to‐3D avatar generation has shown that diffusion models trained on general objects can capture head structure. However, image‐to‐3D avatar that creates a... |
| SourceID | crossref |
| SourceType | Index Database |
| Title | PSHead: 3D Head Reconstruction from a Single Image with Diffusion Prior and Self‐Enhancement |
| WOSCitedRecordID | wos001584839800001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| hasFullText | 1 |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| journalDatabaseRights | – providerCode: PRVWIB databaseName: Wiley Online Library Full Collection 2020 customDbUrl: eissn: 1467-8659 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0004765 issn: 0167-7055 databaseCode: DRFUL dateStart: 19970101 isFulltext: true titleUrlDefault: https://onlinelibrary.wiley.com providerName: Wiley-Blackwell |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV1La9tAEF4SO4f0UJqkJX2ylNyEgvWydnsrdYxbQjCxHUIOMavVrm1wZJNHSW_9Cf2N_SWd0a5WakMhPfSyCCGNZX3DzOzo228JOeBMhJIz6ausE_mxCrt-lnHh51GYqABKaFGyCc-O05MTdn7Oh5Zue1NuJ5AWBbu_5-v_CjWcA7Bx6ew_wO2Mwgk4BtBhBNhhfBTww9EAcYOZftTDVUa4_lCuaqFYs6JEeCNIWkvlfb5C1k7Zju0ttL7D7pk3vF5YcuVILbUjRBwVc3QSR5apFA7szhBeqX6Nss-1xAMGFNuS_lJlScwBd8ZRBFqsM-EM_o_hEXxbOo-7mFvacB8MgY9ZK7ZVESaO9Oa6lxCVUb7HJB8TcTFSs66RBf9LPJczfZjC_JnXSav6UP9HLnMMw2puA7dOy1s3STtME85apN077U-O68WzaTepBODx2az8FNK93O82ipZG9TF-Rp7aaQP9aODeIRuq2CVPGmKSe-TSAP-BRj2KB_R32CnCTgU1sNMSdoqwUwc7LWGnADtF2H9-_9EA_DmZ9I_Gnwa-3T3Dl1BS3_qaKR7nSZgplYggh9pMlJ-xpY5hEi8DmWdcZbh9KNSMmnfykLNYB0qLyIwvSKtYFWqf0DRUHSajWMB1cReCtuR5oEUqZAA2RfSSvK9ez3RtRFKmDwB49ZiLXpPt2nHekBa8I_WWbMmvt4ub63cWul_nG1wx |
| linkProvider | Wiley-Blackwell |
| openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=PSHead%3A+3D+Head+Reconstruction+from+a+Single+Image+with+Diffusion+Prior+and+Self%E2%80%90Enhancement&rft.jtitle=Computer+graphics+forum&rft.au=Yang%2C+Jing&rft.au=Wu%2C+Tianhan&rft.au=Fogarty%2C+Kyle&rft.au=Zhong%2C+Fangcheng&rft.date=2025-10-01&rft.issn=0167-7055&rft.eissn=1467-8659&rft_id=info:doi/10.1111%2Fcgf.70279&rft.externalDBID=n%2Fa&rft.externalDocID=10_1111_cgf_70279 |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0167-7055&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0167-7055&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0167-7055&client=summon |