NaturalInversion: Data-Free Image Synthesis Improving Real-World Consistency (2306.16661v1)
Abstract: We introduce NaturalInversion, a novel model inversion-based method to synthesize images that agrees well with the original data distribution without using real data. In NaturalInversion, we propose: (1) a Feature Transfer Pyramid which uses enhanced image prior of the original data by combining the multi-scale feature maps extracted from the pre-trained classifier, (2) a one-to-one approach generative model where only one batch of images are synthesized by one generator to bring the non-linearity to optimization and to ease the overall optimizing process, (3) learnable Adaptive Channel Scaling parameters which are end-to-end trained to scale the output image channel to utilize the original image prior further. With our NaturalInversion, we synthesize images from classifiers trained on CIFAR-10/100 and show that our images are more consistent with original data distribution than prior works by visualization and additional analysis. Furthermore, our synthesized images outperform prior works on various applications such as knowledge distillation and pruning, demonstrating the effectiveness of our proposed method.
- Data-free learning of student networks. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 3514–3522.
- Detection in crowded scenes: One proposal, multiple predictions. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 12214–12223.
- Imagenet: A large-scale hierarchical image database. In 2009 IEEE conference on computer vision and pattern recognition, 248–255. Ieee.
- Data-free adversarial distillation. arXiv preprint arXiv:1912.11006.
- Texture synthesis using convolutional neural networks. Advances in neural information processing systems, 28: 262–270.
- Image style transfer using convolutional neural networks. In Proceedings of the IEEE conference on computer vision and pattern recognition, 2414–2423.
- Generative adversarial nets. Advances in neural information processing systems, 27.
- The knowledge within: Methods for data-free model compression. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 8494–8502.
- Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition, 770–778.
- A Sliced Wasserstein Loss for Neural Texture Synthesis. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 9412–9420.
- Gans trained by a two time-scale update rule converge to a local nash equilibrium. Advances in neural information processing systems, 30.
- Distilling the knowledge in a neural network. arXiv preprint arXiv:1503.02531.
- Batch normalization: Accelerating deep network training by reducing internal covariate shift. In International conference on machine learning, 448–456. PMLR.
- How much position information do convolutional neural networks encode? arXiv preprint arXiv:2001.08248.
- In the light of feature distributions: moment matching for Neural Style Transfer. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 9382–9391.
- Learning multiple layers of features from tiny images. Technical Report 0, University of Toronto, Toronto, Ontario.
- Drafting and Revision: Laplacian Pyramid Network for Fast High-Quality Artistic Style Transfer. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 5141–5150.
- Learning efficient convolutional networks through network slimming. In Proceedings of the IEEE international conference on computer vision, 2736–2744.
- Rethinking the value of network pruning. arXiv preprint arXiv:1810.05270.
- Understanding deep image representations by inverting them. In Proceedings of the IEEE conference on computer vision and pattern recognition, 5188–5196.
- Conditional generative adversarial nets. arXiv preprint arXiv:1411.1784.
- Deepdream-a code example for visualizing neural networks. Google Research, 2(5).
- Feature visualization. Distill, 2(11): e7.
- Assessing generative models via precision and recall. arXiv preprint arXiv:1806.00035.
- Improved techniques for training gans. Advances in neural information processing systems, 29: 2234–2242.
- Mobilenetv2: Inverted residuals and linear bottlenecks. In Proceedings of the IEEE conference on computer vision and pattern recognition, 4510–4520.
- Semantic pyramid for image generation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 7457–7466.
- Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556.
- Rethinking the inception architecture for computer vision. In Proceedings of the IEEE conference on computer vision and pattern recognition, 2818–2826.
- Visualizing data using t-SNE. Journal of machine learning research, 9(11).
- IMAGINE: Image Synthesis by Image-Guided Model Inversion. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 3681–3690.
- Generative hierarchical features from synthesizing images. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 4432–4442.
- Dreaming to distill: Data-free knowledge transfer via deepinversion. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 8715–8724.
- Understanding neural networks through deep visualization. arXiv preprint arXiv:1506.06579.
- Visualizing and understanding convolutional networks. In European conference on computer vision, 818–833. Springer.
- Scaling vision transformers. arXiv preprint arXiv:2106.04560.