Matches in SemOpenAlex for { <https://semopenalex.org/work/W3129408081> ?p ?o ?g. }
Showing items 1 to 69 of
69
with 100 items per page.
- W3129408081 abstract "We investigate the task of entity counting in overhead imagery from the perspective of re-purposing representations learned from ground imagery, e.g., ImageNet, via feature adaptation. We explore two directions of feature adaptation and analyze their performances using two popular aerial datasets for vehicle counting: PUCPR+ and CARPK. First, we explore proxy self-supervision tasks such as RotNet, jigsaw, and image inpainting to re-fine the pretrained representation. Second, we insert additional network layers to adaptively select suitable features (e.g., squeeze and excitation blocks) or impose desired properties (e.g., using active rotating filters for rotation invariance). Our experimental results show that different adaptations produce different amounts of performance improvements depending on data characteristics. Overall, we achieve a mean absolute error (MAE) of 3.71 and 5.93 on the PUCPR+ and CARPK datasets, respectively, outperforming the previous state of the art: MAEs of 5.24 for PUCPR+ and 7.48 for CARPK." @default.
- W3129408081 created "2021-03-01" @default.
- W3129408081 creator A5028124265 @default.
- W3129408081 creator A5038859737 @default.
- W3129408081 creator A5052203401 @default.
- W3129408081 creator A5080330227 @default.
- W3129408081 date "2021-01-01" @default.
- W3129408081 modified "2023-10-18" @default.
- W3129408081 title "Fine-Tuning for One-Look Regression Vehicle Counting in Low-Shot Aerial Datasets" @default.
- W3129408081 cites W2108598243 @default.
- W3129408081 cites W2117539524 @default.
- W3129408081 cites W2171196813 @default.
- W3129408081 cites W2321533354 @default.
- W3129408081 cites W2331128040 @default.
- W3129408081 cites W2336763592 @default.
- W3129408081 cites W2593414223 @default.
- W3129408081 cites W2732026016 @default.
- W3129408081 cites W2752782242 @default.
- W3129408081 cites W2804199516 @default.
- W3129408081 cites W2900358019 @default.
- W3129408081 cites W2913939497 @default.
- W3129408081 cites W2962689769 @default.
- W3129408081 cites W2962749812 @default.
- W3129408081 cites W2962921175 @default.
- W3129408081 cites W2962945654 @default.
- W3129408081 cites W2963231953 @default.
- W3129408081 cites W2963299736 @default.
- W3129408081 cites W2963378109 @default.
- W3129408081 cites W2963826106 @default.
- W3129408081 cites W2964285767 @default.
- W3129408081 cites W2964979676 @default.
- W3129408081 cites W2971900262 @default.
- W3129408081 cites W2987185654 @default.
- W3129408081 cites W2991359031 @default.
- W3129408081 cites W3003588906 @default.
- W3129408081 cites W343636949 @default.
- W3129408081 doi "https://doi.org/10.1007/978-3-030-68793-9_1" @default.
- W3129408081 hasPublicationYear "2021" @default.
- W3129408081 type Work @default.
- W3129408081 sameAs 3129408081 @default.
- W3129408081 citedByCount "0" @default.
- W3129408081 crossrefType "book-chapter" @default.
- W3129408081 hasAuthorship W3129408081A5028124265 @default.
- W3129408081 hasAuthorship W3129408081A5038859737 @default.
- W3129408081 hasAuthorship W3129408081A5052203401 @default.
- W3129408081 hasAuthorship W3129408081A5080330227 @default.
- W3129408081 hasConcept C153180895 @default.
- W3129408081 hasConcept C154945302 @default.
- W3129408081 hasConcept C41008148 @default.
- W3129408081 hasConceptScore W3129408081C153180895 @default.
- W3129408081 hasConceptScore W3129408081C154945302 @default.
- W3129408081 hasConceptScore W3129408081C41008148 @default.
- W3129408081 hasLocation W31294080811 @default.
- W3129408081 hasOpenAccess W3129408081 @default.
- W3129408081 hasPrimaryLocation W31294080811 @default.
- W3129408081 hasRelatedWork W10793771 @default.
- W3129408081 hasRelatedWork W11383935 @default.
- W3129408081 hasRelatedWork W1602910 @default.
- W3129408081 hasRelatedWork W2988963 @default.
- W3129408081 hasRelatedWork W4033802 @default.
- W3129408081 hasRelatedWork W40380 @default.
- W3129408081 hasRelatedWork W4266367 @default.
- W3129408081 hasRelatedWork W5386269 @default.
- W3129408081 hasRelatedWork W9388322 @default.
- W3129408081 hasRelatedWork W9783904 @default.
- W3129408081 isParatext "false" @default.
- W3129408081 isRetracted "false" @default.
- W3129408081 magId "3129408081" @default.
- W3129408081 workType "book-chapter" @default.