Matches in SemOpenAlex for { <https://semopenalex.org/work/W4375854900> ?p ?o ?g. }
- W4375854900 endingPage "103351" @default.
- W4375854900 startingPage "103351" @default.
- W4375854900 abstract "Existing 2D image-based 3D model retrieval (IBMR) methods usually use the pseudo labels as semantic guidance to reduce the domain-wise and class-wise feature distribution difference across annotated 2D images and unlabeled 3D models. However, they cannot entirely guarantee the quality of pseudo labels, which will decrease prediction discriminability and diversity to affect feature distribution alignment. Therefore, we propose a novel unsupervised self-training correction learning (USTCL) network for the IBMR task. Specifically, we first utilize a CNN to encode 2D images and 3D models (described as multi-view images). Then, we design a noise-corrected self-training learning module (NCST) to denoise pseudo labels in an adversarial manner to make the predicted categories more easily discriminated to improve prediction discriminability. Besides, we employ a target-guided pseudo label refining strategy (TPLR) to progressively refine generated pseudo labels to prevent minority categories from being pushed into majority categories, thereby enhancing prediction diversity. Comprehensive experiments on popular IBMR benchmarks validate the effectiveness and robustness of USTCL, e.g., it can achieve the average gains of 54.20%/22.30%, 62.17%/43.21%, 54.21%/31.18%, 63.48%/43.28%, 59.87%/44.49% in terms of NN, FT, ST, F-measure, DCG and the decrease of 29.50%/35.15% corresponding to ANMRR on MI3DOR (21,000 2D images and 7,690 3D models) and MI3DOR-2 (19,694 2D images and 3,982 3D models), respectively." @default.
- W4375854900 created "2023-05-10" @default.
- W4375854900 creator A5012943082 @default.
- W4375854900 creator A5017353282 @default.
- W4375854900 creator A5024121467 @default.
- W4375854900 creator A5034675261 @default.
- W4375854900 creator A5061046810 @default.
- W4375854900 creator A5081485810 @default.
- W4375854900 date "2023-07-01" @default.
- W4375854900 modified "2023-10-16" @default.
- W4375854900 title "Unsupervised self-training correction learning for 2D image-based 3D model retrieval" @default.
- W4375854900 cites W1644641054 @default.
- W4375854900 cites W2122585444 @default.
- W4375854900 cites W2194775991 @default.
- W4375854900 cites W2211722331 @default.
- W4375854900 cites W2560609797 @default.
- W4375854900 cites W2593414223 @default.
- W4375854900 cites W2616287544 @default.
- W4375854900 cites W2790362762 @default.
- W4375854900 cites W2884771968 @default.
- W4375854900 cites W2895281799 @default.
- W4375854900 cites W2963275094 @default.
- W4375854900 cites W2963697299 @default.
- W4375854900 cites W2976674591 @default.
- W4375854900 cites W2981506675 @default.
- W4375854900 cites W2992687594 @default.
- W4375854900 cites W2997310315 @default.
- W4375854900 cites W2998043922 @default.
- W4375854900 cites W3006536754 @default.
- W4375854900 cites W3023254696 @default.
- W4375854900 cites W3025994431 @default.
- W4375854900 cites W3034706886 @default.
- W4375854900 cites W3035541121 @default.
- W4375854900 cites W3035576098 @default.
- W4375854900 cites W3040060312 @default.
- W4375854900 cites W3049237381 @default.
- W4375854900 cites W3076174210 @default.
- W4375854900 cites W3093133634 @default.
- W4375854900 cites W3110185224 @default.
- W4375854900 cites W3126909200 @default.
- W4375854900 cites W3164880353 @default.
- W4375854900 cites W3167095230 @default.
- W4375854900 cites W3173438721 @default.
- W4375854900 cites W3176053087 @default.
- W4375854900 cites W3178218920 @default.
- W4375854900 cites W3181737292 @default.
- W4375854900 cites W3183988645 @default.
- W4375854900 cites W3196056489 @default.
- W4375854900 cites W3202408237 @default.
- W4375854900 cites W3204469733 @default.
- W4375854900 cites W3205407101 @default.
- W4375854900 cites W3205493274 @default.
- W4375854900 cites W4221050104 @default.
- W4375854900 cites W4221146485 @default.
- W4375854900 cites W4226410746 @default.
- W4375854900 cites W4229052301 @default.
- W4375854900 cites W4233104424 @default.
- W4375854900 doi "https://doi.org/10.1016/j.ipm.2023.103351" @default.
- W4375854900 hasPublicationYear "2023" @default.
- W4375854900 type Work @default.
- W4375854900 citedByCount "0" @default.
- W4375854900 crossrefType "journal-article" @default.
- W4375854900 hasAuthorship W4375854900A5012943082 @default.
- W4375854900 hasAuthorship W4375854900A5017353282 @default.
- W4375854900 hasAuthorship W4375854900A5024121467 @default.
- W4375854900 hasAuthorship W4375854900A5034675261 @default.
- W4375854900 hasAuthorship W4375854900A5061046810 @default.
- W4375854900 hasAuthorship W4375854900A5081485810 @default.
- W4375854900 hasConcept C104317684 @default.
- W4375854900 hasConcept C115961682 @default.
- W4375854900 hasConcept C138885662 @default.
- W4375854900 hasConcept C153180895 @default.
- W4375854900 hasConcept C154945302 @default.
- W4375854900 hasConcept C1667742 @default.
- W4375854900 hasConcept C185592680 @default.
- W4375854900 hasConcept C2776401178 @default.
- W4375854900 hasConcept C41008148 @default.
- W4375854900 hasConcept C41895202 @default.
- W4375854900 hasConcept C55493867 @default.
- W4375854900 hasConcept C63479239 @default.
- W4375854900 hasConceptScore W4375854900C104317684 @default.
- W4375854900 hasConceptScore W4375854900C115961682 @default.
- W4375854900 hasConceptScore W4375854900C138885662 @default.
- W4375854900 hasConceptScore W4375854900C153180895 @default.
- W4375854900 hasConceptScore W4375854900C154945302 @default.
- W4375854900 hasConceptScore W4375854900C1667742 @default.
- W4375854900 hasConceptScore W4375854900C185592680 @default.
- W4375854900 hasConceptScore W4375854900C2776401178 @default.
- W4375854900 hasConceptScore W4375854900C41008148 @default.
- W4375854900 hasConceptScore W4375854900C41895202 @default.
- W4375854900 hasConceptScore W4375854900C55493867 @default.
- W4375854900 hasConceptScore W4375854900C63479239 @default.
- W4375854900 hasFunder F4320321001 @default.
- W4375854900 hasIssue "4" @default.
- W4375854900 hasLocation W43758549001 @default.
- W4375854900 hasOpenAccess W4375854900 @default.
- W4375854900 hasPrimaryLocation W43758549001 @default.
- W4375854900 hasRelatedWork W2033914206 @default.