FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4210768747> ?p ?o ?g. }

• W4210768747 endingPage "23" @default.
• W4210768747 startingPage "1" @default.
• W4210768747 abstract "As a basic component in multimedia applications, object detectors are generally trained on a fixed set of classes that are pre-defined. However, new object classes often emerge after the models are trained in practice. Modern object detectors based on Convolutional Neural Networks (CNN) suffer from catastrophic forgetting when fine-tuning on new classes without the original training data. Therefore, it is critical to improve the incremental learning capability on object detection. In this article, we propose a novel Residual-Distillation-based Incremental learning method on Object Detection (RD-IOD). Our approach rests on the creation of a triple-network based on Faster R-CNN. To enable continuous learning from new classes, we use the original model as well as a residual model to guide the learning of the incremental model on new classes while maintaining the previous learned knowledge. To better maintain the discrimination between the features of old and new classes, the residual model is jointly trained with the incremental model on new classes in the incremental learning procedure. In addition, a two-level distillation scheme is designed to guide the training process, which consists of (1) a general distillation for imitating the original model in feature space along with a residual distillation on the features in both image level and instance level, and (2) a joint classification distillation on the output layers. To well preserve the learned knowledge, we design a 2-threshold training strategy to guide the learning of a Region Proposal Network and a detection head. Extensive experiments conducted on VOC2007 and COCO demonstrate that the proposed method can effectively learn to incrementally detect objects of new classes, and the problem of catastrophic forgetting is mitigated. Our code is available at https://github.com/yangdb/RD-IOD." @default.
• W4210768747 created "2022-02-08" @default.
• W4210768747 creator A5002436544 @default.
• W4210768747 creator A5016175345 @default.
• W4210768747 creator A5016775929 @default.
• W4210768747 creator A5022270828 @default.
• W4210768747 creator A5077114895 @default.
• W4210768747 date "2022-01-27" @default.
• W4210768747 modified "2023-10-16" @default.
• W4210768747 title "RD-IOD: Two-Level Residual-Distillation-Based Triple-Network for Incremental Object Detection" @default.
• W4210768747 cites W1536680647 @default.
• W4210768747 cites W1861492603 @default.
• W4210768747 cites W1991367009 @default.
• W4210768747 cites W2015563892 @default.
• W4210768747 cites W2031489346 @default.
• W4210768747 cites W2060277733 @default.
• W4210768747 cites W2103753221 @default.
• W4210768747 cites W2166344886 @default.
• W4210768747 cites W2194775991 @default.
• W4210768747 cites W2473930607 @default.
• W4210768747 cites W2554616628 @default.
• W4210768747 cites W2560647685 @default.
• W4210768747 cites W2904531787 @default.
• W4210768747 cites W2948734064 @default.
• W4210768747 cites W2950557191 @default.
• W4210768747 cites W2962966271 @default.
• W4210768747 cites W2963018216 @default.
• W4210768747 cites W2963351448 @default.
• W4210768747 cites W2963588172 @default.
• W4210768747 cites W2964189064 @default.
• W4210768747 cites W2964241181 @default.
• W4210768747 cites W2966730026 @default.
• W4210768747 cites W2977932430 @default.
• W4210768747 cites W2983156430 @default.
• W4210768747 cites W2984276908 @default.
• W4210768747 cites W2990154684 @default.
• W4210768747 cites W2997907976 @default.
• W4210768747 cites W3003861315 @default.
• W4210768747 cites W3013325675 @default.
• W4210768747 cites W3034381931 @default.
• W4210768747 cites W3034447740 @default.
• W4210768747 cites W3039204154 @default.
• W4210768747 cites W3103800629 @default.
• W4210768747 cites W3106250896 @default.
• W4210768747 cites W3161907096 @default.
• W4210768747 cites W3178307467 @default.
• W4210768747 cites W7746136 @default.
• W4210768747 doi "https://doi.org/10.1145/3472393" @default.
• W4210768747 hasPublicationYear "2022" @default.
• W4210768747 type Work @default.
• W4210768747 citedByCount "7" @default.
• W4210768747 countsByYear W42107687472022 @default.
• W4210768747 countsByYear W42107687472023 @default.
• W4210768747 crossrefType "journal-article" @default.
• W4210768747 hasAuthorship W4210768747A5002436544 @default.
• W4210768747 hasAuthorship W4210768747A5016175345 @default.
• W4210768747 hasAuthorship W4210768747A5016775929 @default.
• W4210768747 hasAuthorship W4210768747A5022270828 @default.
• W4210768747 hasAuthorship W4210768747A5077114895 @default.
• W4210768747 hasConcept C11413529 @default.
• W4210768747 hasConcept C119857082 @default.
• W4210768747 hasConcept C138885662 @default.
• W4210768747 hasConcept C153180895 @default.
• W4210768747 hasConcept C154945302 @default.
• W4210768747 hasConcept C155512373 @default.
• W4210768747 hasConcept C178790620 @default.
• W4210768747 hasConcept C185592680 @default.
• W4210768747 hasConcept C204030448 @default.
• W4210768747 hasConcept C2776151529 @default.
• W4210768747 hasConcept C2776401178 @default.
• W4210768747 hasConcept C2779542340 @default.
• W4210768747 hasConcept C2781238097 @default.
• W4210768747 hasConcept C41008148 @default.
• W4210768747 hasConcept C41895202 @default.
• W4210768747 hasConcept C7149132 @default.
• W4210768747 hasConcept C81363708 @default.
• W4210768747 hasConceptScore W4210768747C11413529 @default.
• W4210768747 hasConceptScore W4210768747C119857082 @default.
• W4210768747 hasConceptScore W4210768747C138885662 @default.
• W4210768747 hasConceptScore W4210768747C153180895 @default.
• W4210768747 hasConceptScore W4210768747C154945302 @default.
• W4210768747 hasConceptScore W4210768747C155512373 @default.
• W4210768747 hasConceptScore W4210768747C178790620 @default.
• W4210768747 hasConceptScore W4210768747C185592680 @default.
• W4210768747 hasConceptScore W4210768747C204030448 @default.
• W4210768747 hasConceptScore W4210768747C2776151529 @default.
• W4210768747 hasConceptScore W4210768747C2776401178 @default.
• W4210768747 hasConceptScore W4210768747C2779542340 @default.
• W4210768747 hasConceptScore W4210768747C2781238097 @default.
• W4210768747 hasConceptScore W4210768747C41008148 @default.
• W4210768747 hasConceptScore W4210768747C41895202 @default.
• W4210768747 hasConceptScore W4210768747C7149132 @default.
• W4210768747 hasConceptScore W4210768747C81363708 @default.
• W4210768747 hasFunder F4320321001 @default.
• W4210768747 hasFunder F4320325902 @default.
• W4210768747 hasIssue "1" @default.
• W4210768747 hasLocation W42107687471 @default.
• W4210768747 hasOpenAccess W4210768747 @default.

• next