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...
Egile Nagusiak: | Zhang, F, Torr, P, Ranftl, R, Richter, S |
---|---|
Formatua: | Conference item |
Hizkuntza: | English |
Argitaratua: |
Neural Information Processing Systems Foundation
2021
|
Antzeko izenburuak
-
Open vocabulary semantic segmentation with Patch Aligned Contrastive Learning
nork: Mukhoti, J, et al.
Argitaratua: (2023) -
Scalable cascade inference for semantic image segmentation
nork: Sturgess, P, et al.
Argitaratua: (2012) -
Dense semantic image segmentation with objects and attributes
nork: Zheng, S, et al.
Argitaratua: (2014) -
Pyramid Context Contrast for Semantic Segmentation
nork: Yuzhong Chen, et al.
Argitaratua: (2019-01-01) -
Highly Contrast Image Correction for Dim Boundary Separation of Image Semantic Segmentation
nork: Jinyeob Choi, et al.
Argitaratua: (2021-01-01)