Enhance dark images using a learned color invariant and instance-aware translation with DiCo
Enhance dark images using a learned color invariant and instance-aware translation with DiCo
Disentangled Contrastive Image Translation for Nighttime Surveillance
arXiv paper abstract https://arxiv.org/abs/2307.05038
arXiv PDF paper https://arxiv.org/pdf/2307.05038.pdf
Nighttime surveillance suffers from degradation due to poor illumination and arduous human annotations.
... argue that the ultimate solution for nighttime surveillance is night-to-day translation, or Night2Day, which aims to translate a surveillance scene from nighttime to the daytime while maintaining semantic consistency.
To achieve this, this paper presents a Disentangled Contrastive (DiCo) learning method.
... propose a learnable physical prior, i.e., the color invariant, which provides a stable perception of a highly dynamic night environment and can be incorporated into the learning pipeline of neural networks.
... develop a disentangled representation, which is an auxiliary pretext task that separates surveillance scenes into the foreground and background ... can extract the semantics without supervision and boost ... achieve instance-aware translation.
... demonstrate that ... method outperforms existing works significantly ...
Please like and share this post if you enjoyed it using the buttons at the bottom!
Stay up to date. Subscribe to my posts https://morrislee1234.wixsite.com/website/contact
Web site with my other posts by category https://morrislee1234.wixsite.com/website
Comentários