Looking beyond single images for contrastive semantic segmentation learning
We present an approach to contrastive representation learning for semantic segmentation. Our approach leverages the representational power of existing feature extractors to find corresponding regions across images. These cross-image correspondences are used as auxiliary labels to guide the pixel-lev...
Asıl Yazarlar: | Zhang, F, Torr, P, Ranftl, R, Richter, S |
---|---|
Materyal Türü: | Conference item |
Dil: | English |
Baskı/Yayın Bilgisi: |
Neural Information Processing Systems Foundation
2021
|
Benzer Materyaller
-
Open vocabulary semantic segmentation with Patch Aligned Contrastive Learning
Yazar:: Mukhoti, J, ve diğerleri
Baskı/Yayın Bilgisi: (2023) -
Scalable cascade inference for semantic image segmentation
Yazar:: Sturgess, P, ve diğerleri
Baskı/Yayın Bilgisi: (2012) -
Dense semantic image segmentation with objects and attributes
Yazar:: Zheng, S, ve diğerleri
Baskı/Yayın Bilgisi: (2014) -
Pyramid Context Contrast for Semantic Segmentation
Yazar:: Yuzhong Chen, ve diğerleri
Baskı/Yayın Bilgisi: (2019-01-01) -
Highly Contrast Image Correction for Dim Boundary Separation of Image Semantic Segmentation
Yazar:: Jinyeob Choi, ve diğerleri
Baskı/Yayın Bilgisi: (2021-01-01)