densechen / AReLU
AReLU: Attention-based-Rectified-Linear-Unit
☆61Updated 3 years ago
Related projects: ⓘ
- Reducing Channel Redundancy in Convolutional Neural Networks by Features Recombining (TIP 2021)☆18Updated last year
- [ECCV 2020] Scale Adaptive Network: Learning to Learn Parameterized Classification Networks for Scalable Input Images☆43Updated 4 years ago
- The Pytorch implementation of "Location-aware Upsampling for Semantic Segmentation" (LaU)☆52Updated 4 years ago
- Advanced optimizer with Gradient-Centralization☆21Updated 4 years ago
- code and trained models for "Attention as Activation"☆19Updated 4 years ago
- Unofficial PyTorch implementation of "Filter Response Normalization Layer: Eliminating Batch Dependence in the Training of Deep Neural Ne…☆22Updated 4 years ago
- ☆25Updated 4 years ago
- Implementation for our paper exploring a novel 2D adaptive attention span kernel in computer vision.☆35Updated 11 months ago
- ☆49Updated 4 years ago
- A Pytorch implementation of https://arxiv.org/abs/1810.12348.☆36Updated 5 years ago
- The official implementation of paper "Instance Enhancement Batch Normalization: an Adaptive Regulator of Batch Noise".☆40Updated last year
- ☆45Updated 2 years ago
- [ICIP 2019] : Official PyTorch implementation of the paper "What's There in The Dark" accepted in IEEE International Conference in Image …☆29Updated last year
- Code for meta-learning initializations for image segmentation☆30Updated 3 years ago
- Channelized Axial Attention for Semantic Segmentation (AAAI-2022)☆31Updated 2 years ago
- TF 2 implementation Learning to Resize Images for Computer Vision Tasks (https://arxiv.org/abs/2103.09950v1).☆51Updated 2 years ago
- pytorch implementation of Funnel Activation (FReLU)☆16Updated 4 years ago
- (CVPR-Oral 2021) PyTorch implementation of Knowledge Evolution approach and Split-Nets☆83Updated 2 years ago
- PyTorch implementation of Lambda Network and pretrained Lambda-ResNet☆54Updated 3 years ago
- ☆57Updated 2 years ago
- Scene Segmentation with Dual Relation-aware Attention Network (TNNLS2020)