Matches in SemOpenAlex for { <https://semopenalex.org/work/W4287102860> ?p ?o ?g. }
Showing items 1 to 65 of
65
with 100 items per page.
- W4287102860 abstract "Semantic segmentation is one of the basic, yet essential scene understanding tasks for an autonomous agent. The recent developments in supervised machine learning and neural networks have enjoyed great success in enhancing the performance of the state-of-the-art techniques for this task. However, their superior performance is highly reliant on the availability of a large-scale annotated dataset. In this paper, we propose a novel fully unsupervised semantic segmentation method, the so-called Information Maximization and Adversarial Regularization Segmentation (InMARS). Inspired by human perception which parses a scene into perceptual groups, rather than analyzing each pixel individually, our proposed approach first partitions an input image into meaningful regions (also known as superpixels). Next, it utilizes Mutual-Information-Maximization followed by an adversarial training strategy to cluster these regions into semantically meaningful classes. To customize an adversarial training scheme for the problem, we incorporate adversarial pixel noise along with spatial perturbations to impose photometrical and geometrical invariance on the deep neural network. Our experiments demonstrate that our method achieves the state-of-the-art performance on two commonly used unsupervised semantic segmentation datasets, COCO-Stuff, and Potsdam." @default.
- W4287102860 created "2022-07-25" @default.
- W4287102860 creator A5034608678 @default.
- W4287102860 creator A5082741375 @default.
- W4287102860 creator A5036848175 @default.
- W4287102860 date "2021-07-01" @default.
- W4287102860 modified "2023-09-26" @default.
- W4287102860 title "Unsupervised Image Segmentation by Mutual Information Maximization and Adversarial Regularization" @default.
- W4287102860 doi "https://doi.org/10.48550/arxiv.2107.00691" @default.
- W4287102860 hasPublicationYear "2021" @default.
- W4287102860 type Work @default.
- W4287102860 citedByCount "0" @default.
- W4287102860 crossrefType "posted-content" @default.
- W4287102860 hasAuthorship W4287102860A5034608678 @default.
- W4287102860 hasAuthorship W4287102860A5036848175 @default.
- W4287102860 hasAuthorship W4287102860A5082741375 @default.
- W4287102860 hasBestOaLocation W42871028601 @default.
- W4287102860 hasConcept C108583219 @default.
- W4287102860 hasConcept C119857082 @default.
- W4287102860 hasConcept C124504099 @default.
- W4287102860 hasConcept C126255220 @default.
- W4287102860 hasConcept C153180895 @default.
- W4287102860 hasConcept C154945302 @default.
- W4287102860 hasConcept C160633673 @default.
- W4287102860 hasConcept C184337299 @default.
- W4287102860 hasConcept C199360897 @default.
- W4287102860 hasConcept C2776135515 @default.
- W4287102860 hasConcept C2776330181 @default.
- W4287102860 hasConcept C2984842247 @default.
- W4287102860 hasConcept C33923547 @default.
- W4287102860 hasConcept C37736160 @default.
- W4287102860 hasConcept C41008148 @default.
- W4287102860 hasConcept C89600930 @default.
- W4287102860 hasConceptScore W4287102860C108583219 @default.
- W4287102860 hasConceptScore W4287102860C119857082 @default.
- W4287102860 hasConceptScore W4287102860C124504099 @default.
- W4287102860 hasConceptScore W4287102860C126255220 @default.
- W4287102860 hasConceptScore W4287102860C153180895 @default.
- W4287102860 hasConceptScore W4287102860C154945302 @default.
- W4287102860 hasConceptScore W4287102860C160633673 @default.
- W4287102860 hasConceptScore W4287102860C184337299 @default.
- W4287102860 hasConceptScore W4287102860C199360897 @default.
- W4287102860 hasConceptScore W4287102860C2776135515 @default.
- W4287102860 hasConceptScore W4287102860C2776330181 @default.
- W4287102860 hasConceptScore W4287102860C2984842247 @default.
- W4287102860 hasConceptScore W4287102860C33923547 @default.
- W4287102860 hasConceptScore W4287102860C37736160 @default.
- W4287102860 hasConceptScore W4287102860C41008148 @default.
- W4287102860 hasConceptScore W4287102860C89600930 @default.
- W4287102860 hasLocation W42871028601 @default.
- W4287102860 hasOpenAccess W4287102860 @default.
- W4287102860 hasPrimaryLocation W42871028601 @default.
- W4287102860 hasRelatedWork W2610321374 @default.
- W4287102860 hasRelatedWork W2907667403 @default.
- W4287102860 hasRelatedWork W2948658236 @default.
- W4287102860 hasRelatedWork W2960184797 @default.
- W4287102860 hasRelatedWork W2972093541 @default.
- W4287102860 hasRelatedWork W3115553566 @default.
- W4287102860 hasRelatedWork W3193857078 @default.
- W4287102860 hasRelatedWork W4285827401 @default.
- W4287102860 hasRelatedWork W4293054861 @default.
- W4287102860 hasRelatedWork W4319600986 @default.
- W4287102860 isParatext "false" @default.
- W4287102860 isRetracted "false" @default.
- W4287102860 workType "article" @default.