-
Residual attention network for image classification[paper] [slides]- Fei Wang et al., 2017 CVPR
- stacked attention + residual learning
-
Multimodal Unsupervised Image-to-Image Translation[paper] [slides]- Xun Huang et al., 2018 archive
-
FaceNet : A Unified Embedding for Face Recognition and Clustering[paper] [slides]- Florian Schroff et al., 2015 CVPR
- Embedding face to a 128 dimensional vector
- Triplet Loss
-
DenseASPP for Semantic Segmentation in Street Scenes[paper] [slides]- Maoke Yang et al., 2018 CVPR
- DenseNet + ASPP
- Cityscapes dataset SOTA (78.9 mIoU)
-
Few-Shot Segmentation Propagation with Guided Networks[paper] [slides]- Kate Rakelly et al., 2018 archive
- Few-Shot learning
-
Batch-Instance Normalization for Adaptively Style-Invariant Neural Networks[paper] [slides]- H Nam et al., 2018 archive
-
To learn image super-resolution, use a GAN to learn how to do image degradation first[paper] - A Bulat et al., 2018 ECCV
-
GANimation: Anatomically-aware Facial Animation from a Single Image[paper] - A Pumarola et al., 2018 ECCV
-
Context encoding for semantic segmentation[paper] [slides] - H Zhange et al., 2018 CVPR