Matches in SemOpenAlex for { <https://semopenalex.org/work/W2921985587> ?p ?o ?g. }
- W2921985587 abstract "Given a training dataset composed of images and corresponding category labels, deep convolutional neural networks show a strong ability in mining discriminative parts for image classification. However, deep convolutional neural networks trained with image level labels only tend to focus on the most discriminative parts while missing other object parts, which could provide complementary information. In this paper, we approach this problem from a different perspective. We build complementary parts models in a weakly supervised manner to retrieve information suppressed by dominant object parts detected by convolutional neural networks. Given image level labels only, we first extract rough object instances by performing weakly supervised object detection and instance segmentation using Mask R-CNN and CRF-based segmentation. Then we estimate and search for the best parts model for each object instance under the principle of preserving as much diversity as possible. In the last stage, we build a bi-directional long short-term memory (LSTM) network to fuze and encode the partial information of these complementary parts into a comprehensive feature for image classification. Experimental results indicate that the proposed method not only achieves significant improvement over our baseline models, but also outperforms state-of-the-art algorithms by a large margin (6.7%, 2.8%, 5.2% respectively) on Stanford Dogs 120, Caltech-UCSD Birds 2011-200 and Caltech 256." @default.
- W2921985587 created "2019-03-22" @default.
- W2921985587 creator A5036521470 @default.
- W2921985587 creator A5041412266 @default.
- W2921985587 creator A5071959906 @default.
- W2921985587 date "2019-03-07" @default.
- W2921985587 modified "2023-09-23" @default.
- W2921985587 title "Weakly Supervised Complementary Parts Models for Fine-Grained Image Classification from the Bottom Up" @default.
- W2921985587 cites W1496650988 @default.
- W2921985587 cites W1576445103 @default.
- W2921985587 cites W1787224781 @default.
- W2921985587 cites W1849277567 @default.
- W2921985587 cites W2031489346 @default.
- W2921985587 cites W2064675550 @default.
- W2921985587 cites W2079789819 @default.
- W2921985587 cites W2097117768 @default.
- W2921985587 cites W2102605133 @default.
- W2921985587 cites W2117539524 @default.
- W2921985587 cites W2120419212 @default.
- W2921985587 cites W2129999749 @default.
- W2921985587 cites W2130706954 @default.
- W2921985587 cites W2155893237 @default.
- W2921985587 cites W2161236525 @default.
- W2921985587 cites W2163605009 @default.
- W2921985587 cites W2183341477 @default.
- W2921985587 cites W2194775991 @default.
- W2921985587 cites W2287418003 @default.
- W2921985587 cites W2295107390 @default.
- W2921985587 cites W2306952455 @default.
- W2921985587 cites W2438305798 @default.
- W2921985587 cites W2559348937 @default.
- W2921985587 cites W2565639579 @default.
- W2921985587 cites W2581199236 @default.
- W2921985587 cites W2591924527 @default.
- W2921985587 cites W2613718673 @default.
- W2921985587 cites W2673818281 @default.
- W2921985587 cites W2737725206 @default.
- W2921985587 cites W2738853914 @default.
- W2921985587 cites W2741910023 @default.
- W2921985587 cites W2773003563 @default.
- W2921985587 cites W2776581970 @default.
- W2921985587 cites W2797977484 @default.
- W2921985587 cites W2798365843 @default.
- W2921985587 cites W2798381792 @default.
- W2921985587 cites W2798715809 @default.
- W2921985587 cites W2807931652 @default.
- W2921985587 cites W2883502031 @default.
- W2921985587 cites W2949117887 @default.
- W2921985587 cites W2950073423 @default.
- W2921985587 cites W2962747122 @default.
- W2921985587 cites W2962835968 @default.
- W2921985587 cites W2963001329 @default.
- W2921985587 cites W2963150697 @default.
- W2921985587 cites W2963300078 @default.
- W2921985587 cites W2963697527 @default.
- W2921985587 cites W2963749936 @default.
- W2921985587 cites W2964186069 @default.
- W2921985587 cites W56385144 @default.
- W2921985587 cites W14333344 @default.
- W2921985587 doi "https://doi.org/10.48550/arxiv.1903.02827" @default.
- W2921985587 hasPublicationYear "2019" @default.
- W2921985587 type Work @default.
- W2921985587 sameAs 2921985587 @default.
- W2921985587 citedByCount "0" @default.
- W2921985587 crossrefType "posted-content" @default.
- W2921985587 hasAuthorship W2921985587A5036521470 @default.
- W2921985587 hasAuthorship W2921985587A5041412266 @default.
- W2921985587 hasAuthorship W2921985587A5071959906 @default.
- W2921985587 hasBestOaLocation W29219855871 @default.
- W2921985587 hasConcept C108583219 @default.
- W2921985587 hasConcept C115961682 @default.
- W2921985587 hasConcept C119857082 @default.
- W2921985587 hasConcept C120665830 @default.
- W2921985587 hasConcept C121332964 @default.
- W2921985587 hasConcept C138885662 @default.
- W2921985587 hasConcept C153180895 @default.
- W2921985587 hasConcept C154945302 @default.
- W2921985587 hasConcept C192209626 @default.
- W2921985587 hasConcept C2776151529 @default.
- W2921985587 hasConcept C2776401178 @default.
- W2921985587 hasConcept C2781238097 @default.
- W2921985587 hasConcept C41008148 @default.
- W2921985587 hasConcept C41895202 @default.
- W2921985587 hasConcept C52622490 @default.
- W2921985587 hasConcept C75294576 @default.
- W2921985587 hasConcept C774472 @default.
- W2921985587 hasConcept C81363708 @default.
- W2921985587 hasConcept C89600930 @default.
- W2921985587 hasConcept C97931131 @default.
- W2921985587 hasConceptScore W2921985587C108583219 @default.
- W2921985587 hasConceptScore W2921985587C115961682 @default.
- W2921985587 hasConceptScore W2921985587C119857082 @default.
- W2921985587 hasConceptScore W2921985587C120665830 @default.
- W2921985587 hasConceptScore W2921985587C121332964 @default.
- W2921985587 hasConceptScore W2921985587C138885662 @default.
- W2921985587 hasConceptScore W2921985587C153180895 @default.
- W2921985587 hasConceptScore W2921985587C154945302 @default.
- W2921985587 hasConceptScore W2921985587C192209626 @default.
- W2921985587 hasConceptScore W2921985587C2776151529 @default.
- W2921985587 hasConceptScore W2921985587C2776401178 @default.