Beomyoung Kim

beom

[Beomyoung Kim @BeomyoungKim]

🚀Introducing ZIM: Zero-Shot Image Matting – A Step Beyond SAM (Segment Anything)!🚀

Please visit our project page: https://t.co/tB7aBJQ3U7

#ai #computervision #deeplearning #segmentation #matting

[Beomyoung Kim @BeomyoungKim]

We present our research paper #CVPR2023, The Devil is in the Points: Weakly Semi-Supervised Instance Segmentation via Point-Guided Mask Representation.

Paper: https://t.co/mnCaHVQGKp
Code: https://t.co/zY50gyI61P
Presentation: https://t.co/myBP6Kz6gX

[Beomyoung Kim @BeomyoungKim]

I'm happy to announce that one paper has been accepted to CVPR 2023. Following the last year, this is my second CVPR paper! The paper is about weakly and semi-supervised learning for instance segmentation. The arxiv paper and code will be available soon. Stay tuned!
#CVPR2023

[Beomyoung Kim @BeomyoungKim]

We are pleased to share that our work has been accepted at CVPR 2022.

We propose a new method for weakly-supervised instance segmentation, which aims to reduce the expensive annotation of segmentation using cheaper weak labels.

paper: https://t.co/rbs77ArIwT

#CVPR2022