FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4328123569> ?p ?o ?g. }

• W4328123569 endingPage "1661" @default.
• W4328123569 startingPage "1642" @default.
• W4328123569 abstract "Person search aims to simultaneously localize and identify a query person from uncropped images. To achieve this goal, state-of-the-art models typically add a re-id branch upon two-stage detectors like Faster R-CNN. Owing to the ROI-Align operation, this pipeline yields promising accuracy as re-id features are explicitly aligned with the corresponding object regions, but in the meantime, it introduces high computational overhead due to dense object anchors. In this work, we present an anchor-free approach to efficiently tackling this challenging task, by introducing the following dedicated designs. First, we select an anchor-free detector (i.e., FCOS) as the prototype of our framework. Due to the lack of dense object anchors, it exhibits significantly higher efficiency compared with existing person search models. Second, when directly accommodating this anchor-free detector for person search, there exist several misalignment issues in different levels (i.e., scale, region, and task). To address these issues, we propose an aligned feature aggregation module to generate more discriminative and robust feature embeddings. Accordingly, we name our framework as Feature-Aligned Person Search Network (AlignPS). Third, by investigating the advantages of both anchor-based and anchor-free models, we further augment AlignPS with an ROI-Align head, which significantly improves the robustness of re-id features while still keeping our model highly efficient. Our framework not only achieves state-of-the-art or competitive performance on two challenging person search benchmarks, but can be also extended to other challenging searching tasks such as animal and object search. All the source codes, data, and trained models are available at: https://github.com/daodaofr/alignps ." @default.
• W4328123569 created "2023-03-22" @default.
• W4328123569 creator A5006537883 @default.
• W4328123569 creator A5019092879 @default.
• W4328123569 creator A5019708391 @default.
• W4328123569 creator A5048907728 @default.
• W4328123569 creator A5061215231 @default.
• W4328123569 creator A5089646011 @default.
• W4328123569 date "2023-03-21" @default.
• W4328123569 modified "2023-10-16" @default.
• W4328123569 title "Efficient Person Search: An Anchor-Free Approach" @default.
• W4328123569 cites W1518138188 @default.
• W4328123569 cites W1928419358 @default.
• W4328123569 cites W1979260620 @default.
• W4328123569 cites W2108598243 @default.
• W4328123569 cites W2149350210 @default.
• W4328123569 cites W2151103935 @default.
• W4328123569 cites W2156547346 @default.
• W4328123569 cites W2194775991 @default.
• W4328123569 cites W2256680489 @default.
• W4328123569 cites W2473930607 @default.
• W4328123569 cites W2565639579 @default.
• W4328123569 cites W2570343428 @default.
• W4328123569 cites W2586899202 @default.
• W4328123569 cites W2594507094 @default.
• W4328123569 cites W2601564443 @default.
• W4328123569 cites W2606377603 @default.
• W4328123569 cites W2750432752 @default.
• W4328123569 cites W2798429327 @default.
• W4328123569 cites W2884534488 @default.
• W4328123569 cites W2895675617 @default.
• W4328123569 cites W2899309051 @default.
• W4328123569 cites W2931882224 @default.
• W4328123569 cites W2943826729 @default.
• W4328123569 cites W2952787292 @default.
• W4328123569 cites W2962721361 @default.
• W4328123569 cites W2963037989 @default.
• W4328123569 cites W2963047834 @default.
• W4328123569 cites W2963179609 @default.
• W4328123569 cites W2963351448 @default.
• W4328123569 cites W2963404857 @default.
• W4328123569 cites W2963438548 @default.
• W4328123569 cites W2963574614 @default.
• W4328123569 cites W2963639915 @default.
• W4328123569 cites W2963786238 @default.
• W4328123569 cites W2963787027 @default.
• W4328123569 cites W2963842104 @default.
• W4328123569 cites W2963901085 @default.
• W4328123569 cites W2964241181 @default.
• W4328123569 cites W2966926453 @default.
• W4328123569 cites W2981393440 @default.
• W4328123569 cites W2982247743 @default.
• W4328123569 cites W2982770724 @default.
• W4328123569 cites W2984016429 @default.
• W4328123569 cites W2986357608 @default.
• W4328123569 cites W2987014658 @default.
• W4328123569 cites W2989604896 @default.
• W4328123569 cites W2990501368 @default.
• W4328123569 cites W2997268173 @default.
• W4328123569 cites W3007026795 @default.
• W4328123569 cites W3012573144 @default.
• W4328123569 cites W3034230918 @default.
• W4328123569 cites W3034308694 @default.
• W4328123569 cites W3034879391 @default.
• W4328123569 cites W3035228163 @default.
• W4328123569 cites W3035473155 @default.
• W4328123569 cites W3042857285 @default.
• W4328123569 cites W3043178985 @default.
• W4328123569 cites W3092665420 @default.
• W4328123569 cites W3099168143 @default.
• W4328123569 cites W3102701618 @default.
• W4328123569 cites W3106250896 @default.
• W4328123569 cites W3110290422 @default.
• W4328123569 cites W3117037827 @default.
• W4328123569 cites W3135741259 @default.
• W4328123569 cites W3153681304 @default.
• W4328123569 cites W3170005157 @default.
• W4328123569 cites W3173909755 @default.
• W4328123569 cites W3175630131 @default.
• W4328123569 cites W3176336223 @default.
• W4328123569 cites W3186064588 @default.
• W4328123569 cites W3201335952 @default.
• W4328123569 cites W3215919575 @default.
• W4328123569 cites W639708223 @default.
• W4328123569 doi "https://doi.org/10.1007/s11263-023-01772-3" @default.
• W4328123569 hasPublicationYear "2023" @default.
• W4328123569 type Work @default.
• W4328123569 citedByCount "0" @default.
• W4328123569 crossrefType "journal-article" @default.
• W4328123569 hasAuthorship W4328123569A5006537883 @default.
• W4328123569 hasAuthorship W4328123569A5019092879 @default.
• W4328123569 hasAuthorship W4328123569A5019708391 @default.
• W4328123569 hasAuthorship W4328123569A5048907728 @default.
• W4328123569 hasAuthorship W4328123569A5061215231 @default.
• W4328123569 hasAuthorship W4328123569A5089646011 @default.
• W4328123569 hasBestOaLocation W43281235692 @default.
• W4328123569 hasConcept C104317684 @default.
• W4328123569 hasConcept C111919701 @default.

• next