An Explicit Local and Global Representation Disentanglement Framework with Applications in Deep Clustering and Unsupervised Object Detection

Abstract

Visual data can be understood at different levels of granularity, where global features correspond to semantic-level information and local features correspond to texture patterns. In this work, we propose a framework, called SPLIT, which allows us to disentangle local and global information into two separate sets of latent variables within the variational autoencoder (VAE) framework. Our framework adds generative assumption to the VAE by requiring a subset of the latent variables to generate an auxiliary set of observable data. This additional generative assumption primes the latent variables to local information and encourages the other latent variables to represent global information. We examine three different flavours of VAEs with different generative assumptions. We show that the framework can effectively disentangle local and global information within these models leads to improved representation, with better clustering and unsupervised object detection benchmarks. Finally, we establish connections between SPLIT and recent research in cognitive neuroscience regarding the disentanglement in human visual perception. The code for our experiments is at https://github.com/51616/split-vae .

13 pages and 9 figures

References in corpus (13)

• LSTM: A Search Space Odyssey
• Categorical Reparameterization with Gumbel-Softmax
• Conditional Image Generation with PixelCNN Decoders
• Deep Neural Networks Reveal a Gradient in the Complexity of Neural Representations across the Brain's Ventral Visual Pathway
• The Concrete Distribution: A Continuous Relaxation of Discrete Random Variables
• Deep Unsupervised Clustering with Gaussian Mixture Variational Autoencoders
• Learning Discrete Representations via Information Maximizing Self-Augmented Training
• Measuring the tendency of CNNs to Learn Surface Statistical Regularities
• Transferring End-to-End Visuomotor Control from Simulation to Real World for a Multi-Stage Task
• The Consciousness Prior
• Unsupervised Learning of Object Landmarks through Conditional Image Generation
• Learning Hierarchical Features from Generative Models
• An Explicitly Relational Neural Network Architecture

Cited by in corpus (4)

• Identifying Mislabeled Images in Supervised Learning Utilizing Autoencoder
• Decomposing Normal and Abnormal Features of Medical Images into Discrete Latent Codes for Content-Based Image Retrieval
• GMAIR: Unsupervised Object Detection Based on Spatial Attention and Gaussian Mixture
• Understanding in Artificial Intelligence