• Publications
  • Influence
Squeeze-and-Excitation Networks
TLDR
This work proposes a novel architectural unit, which is term the “Squeeze-and-Excitation” (SE) block, that adaptively recalibrates channel-wise feature responses by explicitly modelling interdependencies between channels and shows that these blocks can be stacked together to form SENet architectures that generalise extremely effectively across different datasets.
Squeeze-and-Excitation Networks
TLDR
This work proposes a novel architectural unit, which is term the "Squeeze-and-Excitation" (SE) block, that adaptively recalibrates channel-wise feature responses by explicitly modelling interdependencies between channels and finds that SE blocks produce significant performance improvements for existing state-of-the-art deep architectures at minimal additional computational cost.
Gather-Excite: Exploiting Feature Context in Convolutional Neural Networks
TLDR
This work proposes a simple, lightweight solution to the issue of limited context propagation in ConvNets, which propagates context across a group of neurons by aggregating responses over their extent and redistributing the aggregates back through the group.
A Key Volume Mining Deep Framework for Action Recognition
TLDR
A key volume mining deep framework to identify key volumes and conduct classification simultaneously and an effective yet simple "unsupervised key volume proposal" method for high quality volume sampling are proposed.
Involution: Inverting the Inherence of Convolution for Visual Recognition
TLDR
The proposed involution operator could be leveraged as fundamental bricks to build the new generation of neural networks for visual recognition, powering different deep learning models on several prevalent benchmarks, including ImageNet classification, COCO detection and segmentation, together with Cityscapes segmentation.
Face Sketch Synthesis by Multidomain Adversarial Learning
TLDR
This paper presents a novel face sketch synthesis method by multidomain adversarial learning (termed MDAL), which overcomes the defects of blurs and deformations toward high-quality synthesis.
Image-to-image Translation via Hierarchical Style Disentanglement
TLDR
Hierarchical Style Disentanglement is proposed, which organizes the labels into a hierarchical tree structure, in which independent tags, exclusive attributes, and disentangled styles are allocated from top to bottom.
ISTR: End-to-End Instance Segmentation with Transformers
TLDR
This paper proposes an instance segmentation Transformer, termed ISTR, which is the first end-to-end framework of its kind, and demonstrates state-of-the-art performance even with approximation-based suboptimal embeddings.
Information Competing Process for Learning Diversified Representations
TLDR
Experiments on image classification and image reconstruction tasks demonstrate the great potential of ICP to learn discriminative and disentangled representations in both supervised and self-supervised learning settings.
Attribute Guided Unpaired Image-to-Image Translation with Semi-supervised Learning
TLDR
An Attribute Guided UIT model termed AGUIT is proposed to tackle multi-modal and multi-domain tasks of UIT jointly with a novel semi-supervised setting, which also merits in representation disentanglement and fine control of outputs.
...
...