Deep Network Interpolation for Continuous Imagery Effect Transition
This provides a universal method for users to adjust visual outputs in applications like image restoration and style transfer, though it is incremental as it builds on existing network interpolation concepts.
The paper tackles the problem of achieving continuous transitions between different imagery effects in low-level vision tasks by proposing Deep Network Interpolation (DNI), which uses linear interpolation in network parameter spaces to enable smooth control without task-specific designs.
Deep convolutional neural network has demonstrated its capability of learning a deterministic mapping for the desired imagery effect. However, the large variety of user flavors motivates the possibility of continuous transition among different output effects. Unlike existing methods that require a specific design to achieve one particular transition (e.g., style transfer), we propose a simple yet universal approach to attain a smooth control of diverse imagery effects in many low-level vision tasks, including image restoration, image-to-image translation, and style transfer. Specifically, our method, namely Deep Network Interpolation (DNI), applies linear interpolation in the parameter space of two or more correlated networks. A smooth control of imagery effects can be achieved by tweaking the interpolation coefficients. In addition to DNI and its broad applications, we also investigate the mechanism of network interpolation from the perspective of learned filters.