An Interpretable Generative Model for Handwritten Digits Synthesis

• Published in: Proceedings - International Conference on Image Processing pp. 1910 - 1914
• Main Authors: Zhu, Yao, Suri, Saksham, Kulkarni, Pranav, Chen, Yueru, Duan, Jiali, Kuo, C.-C. Jay
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.09.2019
• Subjects: Decoding; explainable machine learning; feedforward Design; Generative model; Image reconstruction; Image synthesis; Kernel; Principal component analysis; principal component analysis (PCA); Training; Transforms; variational autoencoder (VAE)
• ISSN: 2381-8549
• DOI: 10.1109/ICIP.2019.8803129

An interpretable generative model for handwritten digits synthesis is proposed in this work. Modern image generative models such as the variational autoencoder (VAE) are trained by backpropagation (BP). The training process is complex, and its underlying mechanism is not transparent. Here, we present an explainable generative model using a feedforward design methodology without BP. Being similar to VAEs, it has an encoder and a decoder. For the encoder design, we derive principal-component-analysis-based (PCA-based) transform kernels using the covariance of its inputs. This process converts input images of correlated pixels to uncorrelated spectral components, which play the same role as latent variables in a VAE system. For the decoder design, we convert randomly generated spectral components to synthesized images through the inverse PCA transform. A subject test is conducted to compare the quality of digits generated using the proposed method and the VAE method. They offer comparable perceptual quality yet our model can be obtained at much lower complexity.