Corpus ID: 195750617

Unsupervised Learning of Object Keypoints for Perception and Control

  title={Unsupervised Learning of Object Keypoints for Perception and Control},
  author={T. Kulkarni and A. Gupta and Catalin Ionescu and Sebastian Borgeaud and M. Reynolds and Andrew Zisserman and V. Mnih},
  • T. Kulkarni, A. Gupta, +4 authors V. Mnih
  • Published in NeurIPS 2019
  • Computer Science
  • The study of object representations in computer vision has primarily focused on developing representations that are useful for image classification, object detection, or semantic segmentation as downstream tasks. [...] Key Method Our method learns from raw video frames in a fully unsupervised manner, by transporting learnt image features between video frames using a keypoint bottleneck. The discovered keypoints track objects and object parts across long time-horizons more accurately than recent similar methods…Expand Abstract
    38 Citations
    Relevance-Guided Modeling of Object Dynamics for Reinforcement Learning.
    Unsupervised Object Keypoint Learning using Local Spatial Predictability
    Unsupervised Learning of Object Structure and Dynamics from Videos
    • 29
    • PDF
    Object-Centric Learning with Slot Attention
    • 12
    • PDF
    Transporter Networks: Rearranging the Visual World for Robotic Manipulation
    • 2
    • PDF
    Entity Abstraction in Visual Model-Based Reinforcement Learning
    • 25
    • Highly Influenced
    • PDF


    Unsupervised Video Object Segmentation for Deep Reinforcement Learning
    • 33
    • PDF
    Unsupervised Discovery of Object Landmarks as Structural Representations
    • 74
    • Highly Influential
    • PDF
    Unsupervised Learning of Object Landmarks by Factorized Spatial Embeddings
    • 79
    • PDF
    Multi-Object Representation Learning with Iterative Variational Inference
    • 86
    • PDF
    Deep spatial autoencoders for visuomotor learning
    • 288
    • PDF
    Learning to Refine Object Segments
    • 570
    • PDF
    Unsupervised Learning of Object Landmarks through Conditional Image Generation
    • 68
    • PDF
    Invariant Information Clustering for Unsupervised Image Classification and Segmentation
    • 113
    • PDF
    Discovery of Latent 3D Keypoints via End-to-end Geometric Reasoning
    • 94
    • PDF