PointRend: Image Segmentation as Rendering
Summary Image segmentation based on convolutional neural networks is often based on regular grids. However, because the label map for interior regions of det...
Recent Posts
Rethinking Few-Shot Image Classification: a Good Embedding Is All You Need?
Summary This paper investigated meta-learning for few-shot learning from the perspective of learning a good embedding model. From this understanding, the aut...
Momentum Contrast for Unsupervised Visual Representation Learning.
Summary This paper introduces a new contrastive learning approach – momentum contrast (MoCo). The key ideas are: 1) implement a queue as the dictionary to st...
Decoupled Weight Decay Regularization.
Summary In DL literature, researchers often use weight decay and L2 regularization interchangeably. It’s because they are equivalent when using SGD as the re...
Desining Network Design Space.
Summary This paper introduces a method for designing the design space of deep neural networks. The core idea is to sample network designs from a design space...