Matches in SemOpenAlex for { <https://semopenalex.org/work/W2902878525> ?p ?o ?g. }
Showing items 1 to 83 of
83
with 100 items per page.
- W2902878525 abstract "Computer vision is difficult, partly because the mathematical function connecting input and output data is often complex, fuzzy and thus hard to learn. A currently popular solution is to design a deep neural network and optimize it on a large-scale dataset. However, as the number of parameters increases, the generalization ability is often not guaranteed, e.g., the model can over-fit due to the limited amount of training data, or fail to converge because the desired function is too difficult to learn. This paper presents an effective framework named progressive recurrent learning (PRL). The core idea is similar to curriculum learning which gradually increases the difficulty of training data. We generalize it to a wide range of vision problems that were previously considered less proper to apply curriculum learning. PRL starts with inserting a recurrent prediction scheme, based on the motivation of feeding the prediction of a vision model to the same model iteratively, so that the auxiliary cues contained in it can be exploited to improve the quality of itself. In order to better optimize this framework, we start with providing perfect prediction, i.e., ground-truth, to the second stage, but gradually replace it with the prediction of the first stage. In the final status, the ground-truth information is not needed any more, so that the entire model works on the real data distribution as in the testing process. We apply PRL to two challenging visual recognition tasks, namely, object localization and semantic segmentation, and demonstrate consistent accuracy gain compared to the baseline training strategy, especially in the scenarios of more difficult vision tasks." @default.
- W2902878525 created "2018-12-11" @default.
- W2902878525 creator A5003527952 @default.
- W2902878525 creator A5012409526 @default.
- W2902878525 creator A5037730998 @default.
- W2902878525 creator A5044307720 @default.
- W2902878525 creator A5067478162 @default.
- W2902878525 creator A5075290241 @default.
- W2902878525 creator A5086706224 @default.
- W2902878525 date "2018-11-29" @default.
- W2902878525 modified "2023-09-27" @default.
- W2902878525 title "Progressive Recurrent Learning for Visual Recognition." @default.
- W2902878525 hasPublicationYear "2018" @default.
- W2902878525 type Work @default.
- W2902878525 sameAs 2902878525 @default.
- W2902878525 citedByCount "0" @default.
- W2902878525 crossrefType "posted-content" @default.
- W2902878525 hasAuthorship W2902878525A5003527952 @default.
- W2902878525 hasAuthorship W2902878525A5012409526 @default.
- W2902878525 hasAuthorship W2902878525A5037730998 @default.
- W2902878525 hasAuthorship W2902878525A5044307720 @default.
- W2902878525 hasAuthorship W2902878525A5067478162 @default.
- W2902878525 hasAuthorship W2902878525A5075290241 @default.
- W2902878525 hasAuthorship W2902878525A5086706224 @default.
- W2902878525 hasConcept C111919701 @default.
- W2902878525 hasConcept C119857082 @default.
- W2902878525 hasConcept C134306372 @default.
- W2902878525 hasConcept C14036430 @default.
- W2902878525 hasConcept C146849305 @default.
- W2902878525 hasConcept C154945302 @default.
- W2902878525 hasConcept C177148314 @default.
- W2902878525 hasConcept C2781238097 @default.
- W2902878525 hasConcept C33923547 @default.
- W2902878525 hasConcept C41008148 @default.
- W2902878525 hasConcept C50644808 @default.
- W2902878525 hasConcept C58166 @default.
- W2902878525 hasConcept C78458016 @default.
- W2902878525 hasConcept C86803240 @default.
- W2902878525 hasConcept C89600930 @default.
- W2902878525 hasConcept C98045186 @default.
- W2902878525 hasConceptScore W2902878525C111919701 @default.
- W2902878525 hasConceptScore W2902878525C119857082 @default.
- W2902878525 hasConceptScore W2902878525C134306372 @default.
- W2902878525 hasConceptScore W2902878525C14036430 @default.
- W2902878525 hasConceptScore W2902878525C146849305 @default.
- W2902878525 hasConceptScore W2902878525C154945302 @default.
- W2902878525 hasConceptScore W2902878525C177148314 @default.
- W2902878525 hasConceptScore W2902878525C2781238097 @default.
- W2902878525 hasConceptScore W2902878525C33923547 @default.
- W2902878525 hasConceptScore W2902878525C41008148 @default.
- W2902878525 hasConceptScore W2902878525C50644808 @default.
- W2902878525 hasConceptScore W2902878525C58166 @default.
- W2902878525 hasConceptScore W2902878525C78458016 @default.
- W2902878525 hasConceptScore W2902878525C86803240 @default.
- W2902878525 hasConceptScore W2902878525C89600930 @default.
- W2902878525 hasConceptScore W2902878525C98045186 @default.
- W2902878525 hasLocation W29028785251 @default.
- W2902878525 hasOpenAccess W2902878525 @default.
- W2902878525 hasPrimaryLocation W29028785251 @default.
- W2902878525 hasRelatedWork W1491543565 @default.
- W2902878525 hasRelatedWork W2147368400 @default.
- W2902878525 hasRelatedWork W2439787475 @default.
- W2902878525 hasRelatedWork W2602553746 @default.
- W2902878525 hasRelatedWork W2772691796 @default.
- W2902878525 hasRelatedWork W2786465559 @default.
- W2902878525 hasRelatedWork W2850509156 @default.
- W2902878525 hasRelatedWork W2901300876 @default.
- W2902878525 hasRelatedWork W2903557836 @default.
- W2902878525 hasRelatedWork W2913446087 @default.
- W2902878525 hasRelatedWork W2949808626 @default.
- W2902878525 hasRelatedWork W2987557705 @default.
- W2902878525 hasRelatedWork W2999490157 @default.
- W2902878525 hasRelatedWork W3000156901 @default.
- W2902878525 hasRelatedWork W3005472813 @default.
- W2902878525 hasRelatedWork W3008124487 @default.
- W2902878525 hasRelatedWork W3035486480 @default.
- W2902878525 hasRelatedWork W3095308078 @default.
- W2902878525 hasRelatedWork W3159526778 @default.
- W2902878525 hasRelatedWork W3167352803 @default.
- W2902878525 isParatext "false" @default.
- W2902878525 isRetracted "false" @default.
- W2902878525 magId "2902878525" @default.
- W2902878525 workType "article" @default.