Matches in SemOpenAlex for { <https://semopenalex.org/work/W4309131024> ?p ?o ?g. }
Showing items 1 to 73 of
73
with 100 items per page.
- W4309131024 abstract "We develop a Synthetic Fusion Pyramid Network (SPF-Net) with a scale-aware loss function design for accurate crowd counting. Existing crowd-counting methods assume that the training annotation points were accurate and thus ignore the fact that noisy annotations can lead to large model-learning bias and counting error, especially for counting highly dense crowds that appear far away. To the best of our knowledge, this work is the first to properly handle such noise at multiple scales in end-to-end loss design and thus push the crowd counting state-of-the-art. We model the noise of crowd annotation points as a Gaussian and derive the crowd probability density map from the input image. We then approximate the joint distribution of crowd density maps with the full covariance of multiple scales and derive a low-rank approximation for tractability and efficient implementation. The derived scale-aware loss function is used to train the SPF-Net. We show that it outperforms various loss functions on four public datasets: UCF-QNRF, UCF CC 50, NWPU and ShanghaiTech A-B datasets. The proposed SPF-Net can accurately predict the locations of people in the crowd, despite training on noisy training annotations." @default.
- W4309131024 created "2022-11-23" @default.
- W4309131024 creator A5007240491 @default.
- W4309131024 creator A5012638803 @default.
- W4309131024 creator A5047638290 @default.
- W4309131024 creator A5079301699 @default.
- W4309131024 creator A5080899738 @default.
- W4309131024 date "2022-11-13" @default.
- W4309131024 modified "2023-10-16" @default.
- W4309131024 title "Scale-Aware Crowd Counting Using a Joint Likelihood Density Map and Synthetic Fusion Pyramid Network" @default.
- W4309131024 doi "https://doi.org/10.48550/arxiv.2211.06835" @default.
- W4309131024 hasPublicationYear "2022" @default.
- W4309131024 type Work @default.
- W4309131024 citedByCount "0" @default.
- W4309131024 crossrefType "posted-content" @default.
- W4309131024 hasAuthorship W4309131024A5007240491 @default.
- W4309131024 hasAuthorship W4309131024A5012638803 @default.
- W4309131024 hasAuthorship W4309131024A5047638290 @default.
- W4309131024 hasAuthorship W4309131024A5079301699 @default.
- W4309131024 hasAuthorship W4309131024A5080899738 @default.
- W4309131024 hasBestOaLocation W43091310241 @default.
- W4309131024 hasConcept C115961682 @default.
- W4309131024 hasConcept C119857082 @default.
- W4309131024 hasConcept C121332964 @default.
- W4309131024 hasConcept C124101348 @default.
- W4309131024 hasConcept C14036430 @default.
- W4309131024 hasConcept C142575187 @default.
- W4309131024 hasConcept C153180895 @default.
- W4309131024 hasConcept C154945302 @default.
- W4309131024 hasConcept C2524010 @default.
- W4309131024 hasConcept C2777852691 @default.
- W4309131024 hasConcept C2778755073 @default.
- W4309131024 hasConcept C33923547 @default.
- W4309131024 hasConcept C38652104 @default.
- W4309131024 hasConcept C41008148 @default.
- W4309131024 hasConcept C62520636 @default.
- W4309131024 hasConcept C78458016 @default.
- W4309131024 hasConcept C86803240 @default.
- W4309131024 hasConcept C99498987 @default.
- W4309131024 hasConceptScore W4309131024C115961682 @default.
- W4309131024 hasConceptScore W4309131024C119857082 @default.
- W4309131024 hasConceptScore W4309131024C121332964 @default.
- W4309131024 hasConceptScore W4309131024C124101348 @default.
- W4309131024 hasConceptScore W4309131024C14036430 @default.
- W4309131024 hasConceptScore W4309131024C142575187 @default.
- W4309131024 hasConceptScore W4309131024C153180895 @default.
- W4309131024 hasConceptScore W4309131024C154945302 @default.
- W4309131024 hasConceptScore W4309131024C2524010 @default.
- W4309131024 hasConceptScore W4309131024C2777852691 @default.
- W4309131024 hasConceptScore W4309131024C2778755073 @default.
- W4309131024 hasConceptScore W4309131024C33923547 @default.
- W4309131024 hasConceptScore W4309131024C38652104 @default.
- W4309131024 hasConceptScore W4309131024C41008148 @default.
- W4309131024 hasConceptScore W4309131024C62520636 @default.
- W4309131024 hasConceptScore W4309131024C78458016 @default.
- W4309131024 hasConceptScore W4309131024C86803240 @default.
- W4309131024 hasConceptScore W4309131024C99498987 @default.
- W4309131024 hasLocation W43091310241 @default.
- W4309131024 hasOpenAccess W4309131024 @default.
- W4309131024 hasPrimaryLocation W43091310241 @default.
- W4309131024 hasRelatedWork W1792107727 @default.
- W4309131024 hasRelatedWork W1809875158 @default.
- W4309131024 hasRelatedWork W1968935646 @default.
- W4309131024 hasRelatedWork W1981023409 @default.
- W4309131024 hasRelatedWork W2073244017 @default.
- W4309131024 hasRelatedWork W2112999064 @default.
- W4309131024 hasRelatedWork W2326429803 @default.
- W4309131024 hasRelatedWork W2392271692 @default.
- W4309131024 hasRelatedWork W2745941246 @default.
- W4309131024 hasRelatedWork W3127217315 @default.
- W4309131024 isParatext "false" @default.
- W4309131024 isRetracted "false" @default.
- W4309131024 workType "article" @default.