Computationally Efficient Approaches for Image Style Transfer
This work provides a computationally efficient solution for image style transfer, which is incremental as it builds on existing methods with specific optimizations.
The paper tackled the problem of reducing computational complexity in image style transfer by proposing efficient architectural modifications, achieving reductions in testing time of 26.1%, 39.1%, and 57.1% while maintaining perceptual quality.
In this work, we have investigated various style transfer approaches and (i) examined how the stylized reconstruction changes with the change of loss function and (ii) provided a computationally efficient solution for the same. We have used elegant techniques like depth-wise separable convolution in place of convolution and nearest neighbor interpolation in place of transposed convolution. Further, we have also added multiple interpolations in place of transposed convolution. The results obtained are perceptually similar in quality, while being computationally very efficient. The decrease in the computational complexity of our architecture is validated by the decrease in the testing time by 26.1%, 39.1%, and 57.1%, respectively.