← Search

SungJoon Son

2 accepted papers

2022

MatteFormer: Transformer-Based Image Matting via Prior-Tokens

CVPR 2022poster

In this paper, we propose a transformer-based image matting model called MatteFormer, which takes full advantage of trimap information in the transformer block. Our method first introduces a prior-token which is a global representation of each trimap region (e.g. foreground, background and unknown).…

Cited by 92PDFcodeScholar
2018

Motion Feature Network: Fixed Motion Filter for Action Recognition

ECCV 2018poster

Spatio-temporal representations in frame sequences play an important role in the task of action recognition. Previously, a method of using optical flow as a temporal information in combination with a set of RGB images that contain spatial information has shown great performance enhancement in the ac…

Cited by 159SourcePDFScholar