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...
المؤلفون الرئيسيون: | Zhang, F, Torr, P, Ranftl, R, Richter, S |
---|---|
التنسيق: | Conference item |
اللغة: | English |
منشور في: |
Neural Information Processing Systems Foundation
2021
|
مواد مشابهة
-
Open vocabulary semantic segmentation with Patch Aligned Contrastive Learning
حسب: Mukhoti, J, وآخرون
منشور في: (2023) -
Scalable cascade inference for semantic image segmentation
حسب: Sturgess, P, وآخرون
منشور في: (2012) -
Dense semantic image segmentation with objects and attributes
حسب: Zheng, S, وآخرون
منشور في: (2014) -
Pyramid Context Contrast for Semantic Segmentation
حسب: Yuzhong Chen, وآخرون
منشور في: (2019-01-01) -
Highly Contrast Image Correction for Dim Boundary Separation of Image Semantic Segmentation
حسب: Jinyeob Choi, وآخرون
منشور في: (2021-01-01)