kiyoon / channel_sampling

Official implementation of "Capturing Temporal Information in a Single Frame: Channel Sampling Strategies for Action Recognition", BMVC 2022

• 👋 Hi, I’m Kiyoon Kim
• 🎓️ PhD student @ The University of Edinburgh, UK
• 👀 I’m interested in deep learning in computer vision, specifically Video Action Recognition
• 💻 I make Vim, Neovim, and Tmux plugins to enhance terminal experience.

Recent Work

An Action Is Worth Multiple Words: Handling Ambiguity in Action Recognition

Kiyoon Kim, Davide Moltisanti, Oisin Mac Aodha, Laura Sevilla-Lara
In BMVC 2022. arXiv

Capturing Temporal Information in a Single Frame: Channel Sampling Strategies for Action Recognition

Kiyoon Kim, Shreyank N Gowda, Oisin Mac Aodha, Laura Sevilla-Lara
In BMVC 2022. arXiv

Explore My Plugins

• nvim-treesitter-textobjects (co-maintainer)
• nvim-hand-gesture: Use a webcam to unlock hand gesture control
• Treemux for Tmux: a modern file explorer as Tmux sidebar!
• Jupynium.nvim: The BEST way to use Jupyter Notebook in vim
• Tmuxsend.vim: send any text from Vim to other Tmux panes. Interactive development, file explorer to command line, etc.
• telescope-insert-path.nvim: Neovim's Telescope extension to insert file path on your vim current buffer
• treesitter-indent-object.nvim: Smart and context-aware indent object in neovim
• haskell-scope-highlighting.nvim: Dynamic contextual scope highlighting for Haskell.