Matches in SemOpenAlex for { <https://semopenalex.org/work/W3203129146> ?p ?o ?g. }
Showing items 1 to 79 of
79
with 100 items per page.
- W3203129146 endingPage "506" @default.
- W3203129146 startingPage "497" @default.
- W3203129146 abstract "For all the ways convolutional neural nets have revolutionized computer vision in recent years, one important aspect has received surprisingly little attention: the effect of image size on the accuracy of tasks being trained for. Typically, to be efficient, the input images are resized to a relatively small spatial resolution (e.g. 224x224), and both training and inference are carried out at this resolution. The actual mechanism for this re-scaling has been an afterthought: Namely, off-the-shelf image resizers such as bilinear and bicubic are commonly used in most machine learning software frameworks. But do these resizers limit the on task performance of the trained networks? The answer is yes. Indeed, we show that the typical linear resizer can be replaced with learned resizers that can substantially improve performance. Importantly, while the classical resizers typically result in better perceptual quality of the downscaled images, our proposed learned resizers do not necessarily give better visual quality, but instead improve task performance. Our learned image resizer is jointly trained with a baseline vision model. This learned CNN-based resizer creates machine friendly visual manipulations that lead to a consistent improvement of the end task metric over the baseline model. Specifically, here we focus on the classification task with the ImageNet dataset, and experiment with four different models to learn resizers adapted to each model. Moreover, we show that the proposed resizer can also be useful for fine-tuning the classification baselines for other vision tasks. To this end, we experiment with three different baselines to develop image quality assessment (IQA) models on the AVA dataset." @default.
- W3203129146 created "2021-10-11" @default.
- W3203129146 creator A5002085979 @default.
- W3203129146 creator A5008529493 @default.
- W3203129146 date "2021-03-17" @default.
- W3203129146 modified "2023-09-23" @default.
- W3203129146 title "Learning To Resize Images for Computer Vision Tasks" @default.
- W3203129146 hasPublicationYear "2021" @default.
- W3203129146 type Work @default.
- W3203129146 sameAs 3203129146 @default.
- W3203129146 citedByCount "0" @default.
- W3203129146 crossrefType "proceedings-article" @default.
- W3203129146 hasAuthorship W3203129146A5002085979 @default.
- W3203129146 hasAuthorship W3203129146A5008529493 @default.
- W3203129146 hasConcept C108583219 @default.
- W3203129146 hasConcept C115961682 @default.
- W3203129146 hasConcept C119857082 @default.
- W3203129146 hasConcept C120665830 @default.
- W3203129146 hasConcept C121332964 @default.
- W3203129146 hasConcept C154945302 @default.
- W3203129146 hasConcept C162324750 @default.
- W3203129146 hasConcept C176217482 @default.
- W3203129146 hasConcept C187736073 @default.
- W3203129146 hasConcept C192209626 @default.
- W3203129146 hasConcept C205203396 @default.
- W3203129146 hasConcept C21547014 @default.
- W3203129146 hasConcept C2776214188 @default.
- W3203129146 hasConcept C2780451532 @default.
- W3203129146 hasConcept C31972630 @default.
- W3203129146 hasConcept C41008148 @default.
- W3203129146 hasConcept C55020928 @default.
- W3203129146 hasConcept C81363708 @default.
- W3203129146 hasConceptScore W3203129146C108583219 @default.
- W3203129146 hasConceptScore W3203129146C115961682 @default.
- W3203129146 hasConceptScore W3203129146C119857082 @default.
- W3203129146 hasConceptScore W3203129146C120665830 @default.
- W3203129146 hasConceptScore W3203129146C121332964 @default.
- W3203129146 hasConceptScore W3203129146C154945302 @default.
- W3203129146 hasConceptScore W3203129146C162324750 @default.
- W3203129146 hasConceptScore W3203129146C176217482 @default.
- W3203129146 hasConceptScore W3203129146C187736073 @default.
- W3203129146 hasConceptScore W3203129146C192209626 @default.
- W3203129146 hasConceptScore W3203129146C205203396 @default.
- W3203129146 hasConceptScore W3203129146C21547014 @default.
- W3203129146 hasConceptScore W3203129146C2776214188 @default.
- W3203129146 hasConceptScore W3203129146C2780451532 @default.
- W3203129146 hasConceptScore W3203129146C31972630 @default.
- W3203129146 hasConceptScore W3203129146C41008148 @default.
- W3203129146 hasConceptScore W3203129146C55020928 @default.
- W3203129146 hasConceptScore W3203129146C81363708 @default.
- W3203129146 hasLocation W32031291461 @default.
- W3203129146 hasOpenAccess W3203129146 @default.
- W3203129146 hasPrimaryLocation W32031291461 @default.
- W3203129146 hasRelatedWork W1489833900 @default.
- W3203129146 hasRelatedWork W2295546024 @default.
- W3203129146 hasRelatedWork W2748182091 @default.
- W3203129146 hasRelatedWork W2798581339 @default.
- W3203129146 hasRelatedWork W2798788395 @default.
- W3203129146 hasRelatedWork W2886658347 @default.
- W3203129146 hasRelatedWork W2896638585 @default.
- W3203129146 hasRelatedWork W2946880267 @default.
- W3203129146 hasRelatedWork W2963289467 @default.
- W3203129146 hasRelatedWork W2990897304 @default.
- W3203129146 hasRelatedWork W2997742385 @default.
- W3203129146 hasRelatedWork W3034427277 @default.
- W3203129146 hasRelatedWork W3048349746 @default.
- W3203129146 hasRelatedWork W3119439276 @default.
- W3203129146 hasRelatedWork W3137956596 @default.
- W3203129146 hasRelatedWork W3174507493 @default.
- W3203129146 hasRelatedWork W3176166976 @default.
- W3203129146 hasRelatedWork W3182485714 @default.
- W3203129146 hasRelatedWork W3198112749 @default.
- W3203129146 hasRelatedWork W3211483028 @default.
- W3203129146 isParatext "false" @default.
- W3203129146 isRetracted "false" @default.
- W3203129146 magId "3203129146" @default.
- W3203129146 workType "article" @default.