Matches in SemOpenAlex for { <https://semopenalex.org/work/W3215749523> ?p ?o ?g. }
- W3215749523 endingPage "107726" @default.
- W3215749523 startingPage "107726" @default.
- W3215749523 abstract "To reduce the manpower consumption on box-level annotations, many weakly supervised object detection methods which only require image-level annotations, have been proposed recently. The training process in these methods is formulated into two steps. They firstly train a neural network under weak supervision to generate pseudo ground truths (PGTs). Then, these PGTs are used to train another network under full supervision. Compared with fully supervised methods, the training process in weakly supervised methods becomes more complex and time-consuming. Furthermore, overwhelming negative proposals are involved at the first step. This is neglected by most methods, which makes the training network biased towards to negative proposals and thus degrades the quality of the PGTs, limiting the training network performance at the second step. Online proposal sampling is an intuitive solution to these issues. However, lacking of adequate labeling, a simple online proposal sampling may make the training network stuck into local minima. To solve this problem, we propose an Online Active Proposal Set Generation (OPG) algorithm. Our OPG algorithm consists of two parts: Dynamic Proposal Constraint (DPC) and Proposal Partition (PP). DPC is proposed to dynamically determine different proposal sampling strategies according to the current training state. PP is used to score each proposal, part proposals into different sets and generate an active proposal set for the network optimization. Through experiments, our proposed OPG shows consistent and significant improvement on both datasets PASCAL VOC 2007 and 2012, yielding comparable performance to the state-of-the-art results." @default.
- W3215749523 created "2021-12-06" @default.
- W3215749523 creator A5017489902 @default.
- W3215749523 creator A5021198455 @default.
- W3215749523 creator A5029912845 @default.
- W3215749523 date "2022-02-01" @default.
- W3215749523 modified "2023-10-18" @default.
- W3215749523 title "Online Active Proposal Set Generation for weakly supervised object detection" @default.
- W3215749523 cites W1536680647 @default.
- W3215749523 cites W1994488211 @default.
- W3215749523 cites W2031489346 @default.
- W3215749523 cites W2088049833 @default.
- W3215749523 cites W2097117768 @default.
- W3215749523 cites W2108598243 @default.
- W3215749523 cites W2194775991 @default.
- W3215749523 cites W2270801692 @default.
- W3215749523 cites W2604260814 @default.
- W3215749523 cites W2813911573 @default.
- W3215749523 cites W2884195989 @default.
- W3215749523 cites W2895236117 @default.
- W3215749523 cites W2924232824 @default.
- W3215749523 cites W2954087924 @default.
- W3215749523 cites W2958020048 @default.
- W3215749523 cites W2962721361 @default.
- W3215749523 cites W2963037989 @default.
- W3215749523 cites W2963516811 @default.
- W3215749523 cites W2963603913 @default.
- W3215749523 cites W2963676873 @default.
- W3215749523 cites W2963949812 @default.
- W3215749523 cites W2963952323 @default.
- W3215749523 cites W2990400263 @default.
- W3215749523 cites W2991662170 @default.
- W3215749523 cites W3004469560 @default.
- W3215749523 cites W3019659794 @default.
- W3215749523 cites W3026303927 @default.
- W3215749523 cites W3087060432 @default.
- W3215749523 cites W3087706302 @default.
- W3215749523 cites W3106250896 @default.
- W3215749523 doi "https://doi.org/10.1016/j.knosys.2021.107726" @default.
- W3215749523 hasPublicationYear "2022" @default.
- W3215749523 type Work @default.
- W3215749523 sameAs 3215749523 @default.
- W3215749523 citedByCount "2" @default.
- W3215749523 countsByYear W32157495232022 @default.
- W3215749523 crossrefType "journal-article" @default.
- W3215749523 hasAuthorship W3215749523A5017489902 @default.
- W3215749523 hasAuthorship W3215749523A5021198455 @default.
- W3215749523 hasAuthorship W3215749523A5029912845 @default.
- W3215749523 hasBestOaLocation W32157495232 @default.
- W3215749523 hasConcept C106131492 @default.
- W3215749523 hasConcept C111919701 @default.
- W3215749523 hasConcept C114614502 @default.
- W3215749523 hasConcept C119857082 @default.
- W3215749523 hasConcept C124101348 @default.
- W3215749523 hasConcept C134306372 @default.
- W3215749523 hasConcept C140779682 @default.
- W3215749523 hasConcept C154945302 @default.
- W3215749523 hasConcept C177264268 @default.
- W3215749523 hasConcept C186633575 @default.
- W3215749523 hasConcept C199360897 @default.
- W3215749523 hasConcept C2781238097 @default.
- W3215749523 hasConcept C31972630 @default.
- W3215749523 hasConcept C33923547 @default.
- W3215749523 hasConcept C41008148 @default.
- W3215749523 hasConcept C42812 @default.
- W3215749523 hasConcept C50644808 @default.
- W3215749523 hasConcept C75608658 @default.
- W3215749523 hasConcept C98045186 @default.
- W3215749523 hasConceptScore W3215749523C106131492 @default.
- W3215749523 hasConceptScore W3215749523C111919701 @default.
- W3215749523 hasConceptScore W3215749523C114614502 @default.
- W3215749523 hasConceptScore W3215749523C119857082 @default.
- W3215749523 hasConceptScore W3215749523C124101348 @default.
- W3215749523 hasConceptScore W3215749523C134306372 @default.
- W3215749523 hasConceptScore W3215749523C140779682 @default.
- W3215749523 hasConceptScore W3215749523C154945302 @default.
- W3215749523 hasConceptScore W3215749523C177264268 @default.
- W3215749523 hasConceptScore W3215749523C186633575 @default.
- W3215749523 hasConceptScore W3215749523C199360897 @default.
- W3215749523 hasConceptScore W3215749523C2781238097 @default.
- W3215749523 hasConceptScore W3215749523C31972630 @default.
- W3215749523 hasConceptScore W3215749523C33923547 @default.
- W3215749523 hasConceptScore W3215749523C41008148 @default.
- W3215749523 hasConceptScore W3215749523C42812 @default.
- W3215749523 hasConceptScore W3215749523C50644808 @default.
- W3215749523 hasConceptScore W3215749523C75608658 @default.
- W3215749523 hasConceptScore W3215749523C98045186 @default.
- W3215749523 hasFunder F4320320751 @default.
- W3215749523 hasFunder F4320320766 @default.
- W3215749523 hasLocation W32157495231 @default.
- W3215749523 hasLocation W32157495232 @default.
- W3215749523 hasOpenAccess W3215749523 @default.
- W3215749523 hasPrimaryLocation W32157495231 @default.
- W3215749523 hasRelatedWork W1965681098 @default.
- W3215749523 hasRelatedWork W1969663039 @default.
- W3215749523 hasRelatedWork W2066229105 @default.
- W3215749523 hasRelatedWork W2079993633 @default.
- W3215749523 hasRelatedWork W2140468882 @default.