Matches in SemOpenAlex for { <https://semopenalex.org/work/W4366308797> ?p ?o ?g. }
- W4366308797 abstract "In recent years, many salient object detection (SOD) methods introduce depth cues to boost detection performance in challenging scenes, named as RGB-D SOD. However, how to effectively fuse cross-modal features with various properties (i.e., RGB and depth) has become a key issue that is hard to be avoided. Most existing methods employ simple operations, such as concatenation or summation, for cross-modal fusion, ignoring the negative effects of low-quality depth maps, thus yielding poor performance. In this paper, we design a simple yet effective fusion method, which utilizes 3D convolution to extract modality-specific and modality-shared information respectively for sufficient cross-modal fusion, and combines modality weights to mitigate the interference of invalid information. In addition, we propose a novel multi-level feature integration strategy in the decoder, which explicitly incorporates the low-level detail information and high-level semantic information into the mid-level to generate accurate saliency maps. Extensive experiments on six public datasets show that our method achieves competitive results compared to 17 state-of-the-art methods." @default.
- W4366308797 created "2023-04-20" @default.
- W4366308797 creator A5016056157 @default.
- W4366308797 creator A5050358356 @default.
- W4366308797 date "2022-10-01" @default.
- W4366308797 modified "2023-10-01" @default.
- W4366308797 title "RGB-D Saliency Detection with 3D Cross-modal Fusion and Mid-level Integration" @default.
- W4366308797 cites W1497443722 @default.
- W4366308797 cites W1966025376 @default.
- W4366308797 cites W1976409045 @default.
- W4366308797 cites W1976754232 @default.
- W4366308797 cites W20683899 @default.
- W4366308797 cites W2086866337 @default.
- W4366308797 cites W2090323693 @default.
- W4366308797 cites W2100470808 @default.
- W4366308797 cites W2133515615 @default.
- W4366308797 cites W2145023731 @default.
- W4366308797 cites W2158535435 @default.
- W4366308797 cites W2314707829 @default.
- W4366308797 cites W2461758788 @default.
- W4366308797 cites W2565639579 @default.
- W4366308797 cites W2765838470 @default.
- W4366308797 cites W2907643346 @default.
- W4366308797 cites W2913602965 @default.
- W4366308797 cites W2948300571 @default.
- W4366308797 cites W2957414648 @default.
- W4366308797 cites W2963529609 @default.
- W4366308797 cites W2963604034 @default.
- W4366308797 cites W2963868681 @default.
- W4366308797 cites W3002301267 @default.
- W4366308797 cites W3035284915 @default.
- W4366308797 cites W3035357085 @default.
- W4366308797 cites W3035633116 @default.
- W4366308797 cites W3035687312 @default.
- W4366308797 cites W3097053213 @default.
- W4366308797 cites W3099871687 @default.
- W4366308797 cites W3104979525 @default.
- W4366308797 cites W3106587394 @default.
- W4366308797 cites W3108421143 @default.
- W4366308797 cites W3108812909 @default.
- W4366308797 cites W3114152269 @default.
- W4366308797 cites W3126725132 @default.
- W4366308797 cites W3173882198 @default.
- W4366308797 cites W3177040887 @default.
- W4366308797 cites W3207804100 @default.
- W4366308797 doi "https://doi.org/10.1109/ictai56018.2022.00201" @default.
- W4366308797 hasPublicationYear "2022" @default.
- W4366308797 type Work @default.
- W4366308797 citedByCount "0" @default.
- W4366308797 crossrefType "proceedings-article" @default.
- W4366308797 hasAuthorship W4366308797A5016056157 @default.
- W4366308797 hasAuthorship W4366308797A5050358356 @default.
- W4366308797 hasConcept C114614502 @default.
- W4366308797 hasConcept C119599485 @default.
- W4366308797 hasConcept C127413603 @default.
- W4366308797 hasConcept C138885662 @default.
- W4366308797 hasConcept C141353440 @default.
- W4366308797 hasConcept C153180895 @default.
- W4366308797 hasConcept C154945302 @default.
- W4366308797 hasConcept C158525013 @default.
- W4366308797 hasConcept C185592680 @default.
- W4366308797 hasConcept C188027245 @default.
- W4366308797 hasConcept C26517878 @default.
- W4366308797 hasConcept C2776401178 @default.
- W4366308797 hasConcept C2780226545 @default.
- W4366308797 hasConcept C31972630 @default.
- W4366308797 hasConcept C33923547 @default.
- W4366308797 hasConcept C38652104 @default.
- W4366308797 hasConcept C41008148 @default.
- W4366308797 hasConcept C41895202 @default.
- W4366308797 hasConcept C45347329 @default.
- W4366308797 hasConcept C50644808 @default.
- W4366308797 hasConcept C71139939 @default.
- W4366308797 hasConcept C82990744 @default.
- W4366308797 hasConcept C87619178 @default.
- W4366308797 hasConceptScore W4366308797C114614502 @default.
- W4366308797 hasConceptScore W4366308797C119599485 @default.
- W4366308797 hasConceptScore W4366308797C127413603 @default.
- W4366308797 hasConceptScore W4366308797C138885662 @default.
- W4366308797 hasConceptScore W4366308797C141353440 @default.
- W4366308797 hasConceptScore W4366308797C153180895 @default.
- W4366308797 hasConceptScore W4366308797C154945302 @default.
- W4366308797 hasConceptScore W4366308797C158525013 @default.
- W4366308797 hasConceptScore W4366308797C185592680 @default.
- W4366308797 hasConceptScore W4366308797C188027245 @default.
- W4366308797 hasConceptScore W4366308797C26517878 @default.
- W4366308797 hasConceptScore W4366308797C2776401178 @default.
- W4366308797 hasConceptScore W4366308797C2780226545 @default.
- W4366308797 hasConceptScore W4366308797C31972630 @default.
- W4366308797 hasConceptScore W4366308797C33923547 @default.
- W4366308797 hasConceptScore W4366308797C38652104 @default.
- W4366308797 hasConceptScore W4366308797C41008148 @default.
- W4366308797 hasConceptScore W4366308797C41895202 @default.
- W4366308797 hasConceptScore W4366308797C45347329 @default.
- W4366308797 hasConceptScore W4366308797C50644808 @default.
- W4366308797 hasConceptScore W4366308797C71139939 @default.
- W4366308797 hasConceptScore W4366308797C82990744 @default.
- W4366308797 hasConceptScore W4366308797C87619178 @default.
- W4366308797 hasFunder F4320321001 @default.
- W4366308797 hasFunder F4320322725 @default.