Description
Spatial Broadcast Decoder is an architecture that aims to improve disentangling, reconstruction accuracy, and generalization to held-out regions in data space. It provides a particularly dramatic benefit when applied to datasets with small objects.
Source: Watters et al.
Image source: Watters et al.
Papers Using This Method
An Interpretable Representation Learning Approach for Diffusion Tensor Imaging2025-05-25AudioSlots: A slot-centric generative model for audio separation2023-05-09Posting Bot Detection on Blockchain-based Social Media Platform using Machine Learning Techniques2020-08-28Superpixel Segmentation via Convolutional Neural Networks with Regularized Information Maximization2020-02-17f-BRS: Rethinking Backpropagating Refinement for Interactive Segmentation2020-01-28Sequence Labeling Approach to the Task of Sentence Boundary Detection2020-01-20An Approach for Time-aware Domain-based Social Influence Prediction2020-01-19Deep Snake for Real-Time Instance Segmentation2020-01-06GENESIS: Generative Scene Inference and Sampling with Object-Centric Latent Representations2019-07-30Omnidirectional Scene Text Detection with Sequential-free Box Discretization2019-06-06Object Instance Annotation With Deep Extreme Level Set Evolution2019-06-01Dynamic Feature Fusion for Semantic Edge Detection2019-02-25MONet: Unsupervised Scene Decomposition and Representation2019-01-22Spatial Broadcast Decoder: A Simple Architecture for Learning Disentangled Representations in VAEs2019-01-21