• Home
  • Search
  • Compositional Scene Representation Learning via Reconstruction: A Survey.
  • Open Access IconOpen Access
  • Cite Icon23
  • https://doi.org/10.1109/tpami.2023.3286184Copy DOI Icon

Compositional Scene Representation Learning via Reconstruction: A Survey.

  • Abstract
  • Literature Map
  • References
  • Citations
  • Similar Papers
Abstract

Visual scenes are composed of visual concepts and have the property of combinatorial explosion. An important reason for humans to efficiently learn from diverse visual scenes is the ability of compositional perception, and it is desirable for artificial intelligence to have similar abilities. Compositional scene representation learning is a task that enables such abilities. In recent years, various methods have been proposed to apply deep neural networks, which have been proven to be advantageous in representation learning, to learn compositional scene representations via reconstruction, advancing this research direction into the deep learning era. Learning via reconstruction is advantageous because it may utilize massive unlabeled data and avoid costly and laborious data annotation. In this survey, we first outline the current progress on reconstruction-based compositional scene representation learning with deep neural networks, including development history and categorizations of existing methods from the perspectives of the modeling of visual scenes and the inference of scene representations; then provide benchmarks, including an open source toolbox to reproduce the benchmark experiments, of representative methods that consider the most extensively studied problem setting and form the foundation for other methods; and finally discuss the limitations of existing methods and future directions of this research topic.

Similar Papers
  • Dissertation

Human-centric 3D representation learning

  • Jan 01, 2025
  • Fangzhou Hong
  • Research Article
  • Citations67

Reconstruction of natural visual scenes from neural spikes with deep neural networks

  • Feb 08, 2020
  • Neural Networks
  • Yichen Zhang +7
  • PDF
  • Research Article
  • Citations113

A wholistic view of continual learning with deep neural networks: Forgotten lessons and the bridge to active and open world learning

  • Jan 20, 2023
  • Neural Networks
  • Martin Mundt +3
  • PDF
  • Research Article
  • Citations65

Representation of visual scenes by local neuronal populations in layer 2/3 of mouse visual cortex

  • Dec 12, 2011
  • Frontiers in Neural Circuits
  • Björn M Kampa +3
  • PDF
  • Research Article
  • Citations18

AlphaILP: thinking visual scenes as differentiable logic programs

  • Mar 14, 2023
  • Machine Learning
  • Hikaru Shindo +3
  • Conference Article
  • Citations39

Multimodal and Crossmodal Representation Learning from Textual and Visual Features with Bidirectional Deep Neural Networks for Video Hyperlinking

  • Oct 16, 2016
  • Vedran Vukotić +2
  • PDF
  • Research Article
  • Citations76

Perceptron: Learning, Generalization, Model Selection, Fault Tolerance, and Role in the Deep Learning Era

  • Dec 13, 2022
  • Mathematics
  • Ke-Lin Du +3
  • Research Article
  • Citations46

Late fusion framework for Acoustic Scene Classification using LPCC, SCMC, and log-Mel band energies with Deep Neural Networks

  • Aug 25, 2020
  • Applied Acoustics
  • Chandrasekhar Paseddula +1
  • Research Article
  • Citations29

Compositional neural scene representations for shading inference

  • Aug 12, 2020
  • ACM Transactions on Graphics
  • Jonathan Granskog +3
  • Dissertation

Improving computational efficiency via tiny machine learning data-centric methods

  • Jan 01, 2025
  • Michael Soricelli
  • PDF
  • Research Article
  • Citations30

Learning Eligibility in Cancer Clinical Trials Using Deep Neural Networks

  • Jul 23, 2018
  • Applied Sciences
  • Aurelia Bustos +1
  • Research Article
  • Citations13

Sememe knowledge computation: a review of recent advances in application and expansion of sememe knowledge bases

  • Apr 29, 2021
  • Frontiers of Computer Science
  • Fanchao Qi +4
  • Research Article
  • Citations60

Graph Embedding Contrastive Multi-Modal Representation Learning for Clustering.

  • Jan 01, 2023
  • IEEE Transactions on Image Processing
  • Wei Xia +4
  • Research Article
  • Citations59

DeepProp: Extracting Deep Features from a Single Image for Edit Propagation

  • May 01, 2016
  • Computer Graphics Forum
  • Yuki Endo +3
  • PDF
  • Research Article
  • Citations10

Bias in Deep Neural Networks in Land Use Characterization for International Development

  • Jul 24, 2021
  • Remote Sensing
  • Do-Hyung Kim +8
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.