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...
Những tác giả chính: | Zhang, F, Torr, P, Ranftl, R, Richter, S |
---|---|
Định dạng: | Conference item |
Ngôn ngữ: | English |
Được phát hành: |
Neural Information Processing Systems Foundation
2021
|
Những quyển sách tương tự
-
Open vocabulary semantic segmentation with Patch Aligned Contrastive Learning
Bằng: Mukhoti, J, et al.
Được phát hành: (2023) -
Scalable cascade inference for semantic image segmentation
Bằng: Sturgess, P, et al.
Được phát hành: (2012) -
Dense semantic image segmentation with objects and attributes
Bằng: Zheng, S, et al.
Được phát hành: (2014) -
Pyramid Context Contrast for Semantic Segmentation
Bằng: Yuzhong Chen, et al.
Được phát hành: (2019-01-01) -
Highly Contrast Image Correction for Dim Boundary Separation of Image Semantic Segmentation
Bằng: Jinyeob Choi, et al.
Được phát hành: (2021-01-01)