Matches in SemOpenAlex for { <https://semopenalex.org/work/W4386065544> ?p ?o ?g. }
Showing items 1 to 91 of
91
with 100 items per page.
- W4386065544 abstract "Compared with previous two-stream trackers, the recent one-stream tracking pipeline, which allows earlier interaction between the template and search region, has achieved a remarkable performance gain. However, existing one-stream trackers always let the template interact with all parts inside the search region throughout all the encoder layers. This could potentially lead to target-background confusion when the extracted feature representations are not sufficiently discriminative. To alleviate this issue, we propose a generalized relation modeling method based on adaptive token division. The proposed method is a generalized formulation of attention-based relation modeling for Transformer tracking, which inherits the merits of both previous two-stream and one-stream pipelines whilst enabling more flexible relation modeling by selecting appropriate search tokens to interact with template tokens. An attention masking strategy and the Gumbel-Softmax technique are introduced to facilitate the parallel computation and end-to-end learning of the token division module. Extensive experiments show that our method is superior to the two-stream and one-stream pipelines and achieves state-of-the-art performance on six challenging benchmarks with a real-time running speed. Code and models are publicly available at https://github.com/Little-Podi/GRM." @default.
- W4386065544 created "2023-08-23" @default.
- W4386065544 creator A5027837326 @default.
- W4386065544 creator A5063248074 @default.
- W4386065544 creator A5071672663 @default.
- W4386065544 date "2023-06-01" @default.
- W4386065544 modified "2023-10-02" @default.
- W4386065544 title "Generalized Relation Modeling for Transformer Tracking" @default.
- W4386065544 cites W2194775991 @default.
- W4386065544 cites W2605173812 @default.
- W4386065544 cites W2799058067 @default.
- W4386065544 cites W2891033863 @default.
- W4386065544 cites W2898200825 @default.
- W4386065544 cites W2962766617 @default.
- W4386065544 cites W2963351448 @default.
- W4386065544 cites W2963534981 @default.
- W4386065544 cites W2964423614 @default.
- W4386065544 cites W2966759264 @default.
- W4386065544 cites W3001584168 @default.
- W4386065544 cites W3035211844 @default.
- W4386065544 cites W3035453691 @default.
- W4386065544 cites W3138516171 @default.
- W4386065544 cites W3167536469 @default.
- W4386065544 cites W3167762749 @default.
- W4386065544 cites W3173344965 @default.
- W4386065544 cites W3204540098 @default.
- W4386065544 cites W3204554907 @default.
- W4386065544 cites W3214586131 @default.
- W4386065544 cites W4214759957 @default.
- W4386065544 cites W4285600988 @default.
- W4386065544 cites W4285603016 @default.
- W4386065544 cites W4304481542 @default.
- W4386065544 cites W4312255167 @default.
- W4386065544 cites W4312472480 @default.
- W4386065544 cites W4312651496 @default.
- W4386065544 cites W4312805142 @default.
- W4386065544 doi "https://doi.org/10.1109/cvpr52729.2023.01792" @default.
- W4386065544 hasPublicationYear "2023" @default.
- W4386065544 type Work @default.
- W4386065544 citedByCount "1" @default.
- W4386065544 countsByYear W43860655442023 @default.
- W4386065544 crossrefType "proceedings-article" @default.
- W4386065544 hasAuthorship W4386065544A5027837326 @default.
- W4386065544 hasAuthorship W4386065544A5063248074 @default.
- W4386065544 hasAuthorship W4386065544A5071672663 @default.
- W4386065544 hasConcept C108583219 @default.
- W4386065544 hasConcept C119599485 @default.
- W4386065544 hasConcept C124101348 @default.
- W4386065544 hasConcept C127413603 @default.
- W4386065544 hasConcept C154945302 @default.
- W4386065544 hasConcept C165801399 @default.
- W4386065544 hasConcept C188441871 @default.
- W4386065544 hasConcept C25343380 @default.
- W4386065544 hasConcept C38652104 @default.
- W4386065544 hasConcept C41008148 @default.
- W4386065544 hasConcept C48145219 @default.
- W4386065544 hasConcept C56461940 @default.
- W4386065544 hasConcept C57501372 @default.
- W4386065544 hasConcept C66322947 @default.
- W4386065544 hasConcept C97931131 @default.
- W4386065544 hasConceptScore W4386065544C108583219 @default.
- W4386065544 hasConceptScore W4386065544C119599485 @default.
- W4386065544 hasConceptScore W4386065544C124101348 @default.
- W4386065544 hasConceptScore W4386065544C127413603 @default.
- W4386065544 hasConceptScore W4386065544C154945302 @default.
- W4386065544 hasConceptScore W4386065544C165801399 @default.
- W4386065544 hasConceptScore W4386065544C188441871 @default.
- W4386065544 hasConceptScore W4386065544C25343380 @default.
- W4386065544 hasConceptScore W4386065544C38652104 @default.
- W4386065544 hasConceptScore W4386065544C41008148 @default.
- W4386065544 hasConceptScore W4386065544C48145219 @default.
- W4386065544 hasConceptScore W4386065544C56461940 @default.
- W4386065544 hasConceptScore W4386065544C57501372 @default.
- W4386065544 hasConceptScore W4386065544C66322947 @default.
- W4386065544 hasConceptScore W4386065544C97931131 @default.
- W4386065544 hasLocation W43860655441 @default.
- W4386065544 hasOpenAccess W4386065544 @default.
- W4386065544 hasPrimaryLocation W43860655441 @default.
- W4386065544 hasRelatedWork W2036982747 @default.
- W4386065544 hasRelatedWork W2099370358 @default.
- W4386065544 hasRelatedWork W2162813111 @default.
- W4386065544 hasRelatedWork W2294602567 @default.
- W4386065544 hasRelatedWork W2743258233 @default.
- W4386065544 hasRelatedWork W2766856689 @default.
- W4386065544 hasRelatedWork W2902321499 @default.
- W4386065544 hasRelatedWork W2915772270 @default.
- W4386065544 hasRelatedWork W4214506192 @default.
- W4386065544 hasRelatedWork W4289242713 @default.
- W4386065544 isParatext "false" @default.
- W4386065544 isRetracted "false" @default.
- W4386065544 workType "article" @default.