There are a variety of attention mechanism techniques that are recently published. Attention based methods are used to focus on specific parts of the input and has been applied in deep learning for different applications like translation, speech recognition, etc.
I am mainly asking about the recent attention based methods for "images", it does not have to be image classification but enough if it is designed for focusing the attention on the ROI in images.
Any ***recent*** publications, suggestions, tutorials will be appreciated.
State of the art, here means: The most recent and effective method/algorithm which beats all the previous methods/techniques/algorithms for a similar task and similar data.
I suggest you follow https://arxiv.org/ftp/arxiv/papers/1811/1811.08513.pdf
https://arxiv.org/ftp/arxiv/papers/1703/1703.00121.pdf
https://towardsdatascience.com/self-attention-in-computer-vision-2782727021f6
Muhammad Ali Thanks a lot Muhammad, these are quite nice resources.
- Badges
- Science topic
- Similar topics
- Linguistics
- Composition Studies
- Publications