Tagger: Deep Unsupervised Perceptual Grouping
arXiv:1606.06724
Abstract
We present a framework for efficient perceptual inference that explicitly reasons about the segmentation of its inputs and features. Rather than being trained for any specific segmentation, our framework learns the grouping process in an unsupervised manner or alongside any supervised task. By enriching the representations of a neural network, we enable it to group the representations of different objects in an iterative manner. By allowing the system to amortize the iterative inference of the groupings, we achieve very fast convergence. In contrast to many other recently proposed methods for addressing multi-object scenes, our system does not assume the inputs to be images and can therefore directly handle other modalities. For multi-digit classification of very cluttered images that require texture segmentation, our method offers improved classification performance over convolutional networks despite being fully connected. Furthermore, we observe that our system greatly improves on the semi-supervised result of a baseline Ladder network on our dataset, indicating that segmentation can also improve sample efficiency.
arXiv comment14 pages + 5 pages supplementary, accepted at NIPS 2016
References in corpus (8)
- Show, Attend and Tell: Neural Image Caption Generation with Visual Attention
- Striving for Simplicity: The All Convolutional Net
- Multi-digit Number Recognition from Street View Imagery using Deep Convolutional Neural Networks
- Deep Generative Stochastic Networks Trainable by Backprop
- Attend, Infer, Repeat: Fast Scene Understanding with Generative Models
- Neuronal Synchrony in Complex-Valued Deep Networks
- Efficient inference in occlusion-aware generative models of images
- Binding via Reconstruction Clustering