SparseMask: Differentiable Connectivity Learning for Dense Image Prediction

Abstract

In this paper, we aim at automatically searching an efficient network architecture for dense image prediction. Particularly, we follow the encoder-decoder style and focus on designing a connectivity structure for the decoder. To achieve that, we design a densely connected network with learnable connections, named Fully Dense Network, which contains a large set of possible final connectivity structures. We then employ gradient descent to search the optimal connectivity from the dense connections. The search process is guided by a novel loss function, which pushes the weight of each connection to be binary and the connections to be sparse. The discovered connectivity achieves competitive results on two segmentation datasets, while runs more than three times faster and requires less than half parameters compared to the state-of-the-art methods. An extensive experiment shows that the discovered connectivity is compatible with various backbones and generalizes well to other dense image prediction tasks.

Accepted by ICCV 2019. Code is available at https://github.com/wuhuikai/SparseMask

References in corpus (12)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Rethinking Atrous Convolution for Semantic Image Segmentation
• Neural Architecture Search with Reinforcement Learning
• MobileNetV2: Inverted Residuals and Linear Bottlenecks
• Path Aggregation Network for Instance Segmentation
• Searching for Efficient Multi-Scale Architectures for Dense Image Prediction
• Holistically-Nested Edge Detection
• Convolutional Neural Fabrics
• Auto-DeepLab: Hierarchical Neural Architecture Search for Semantic Image Segmentation
• Learning a Discriminative Feature Network for Semantic Segmentation
• MaskConnect: Connectivity Learning by Gradient Descent
• ExFuse: Enhancing Feature Fusion for Semantic Segmentation