Proceedings of the International Workshop on Advances in Deep Learning for Image Analysis and Computer Vision (IWADIC 2025)

<Previous Article In Volume
Next Article In Volume>

Research and Analysis of VAE in Image and Data Analysis

Authors
Xuan Sheng1, *
1University of Minnesota, Minneapolis, MN, 55414, USA
*Corresponding author. Email: jodie7121212@gmail.com
Corresponding Author
Xuan Sheng
Available Online 24 April 2026.
DOI
10.2991/978-94-6239-648-7_96How to use a DOI?
Keywords
Variational Autoencoders; Perceptual Loss; Deep Generative Models; Image Reconstruction; Representation Learning
Abstract

Deep generative models find popular applications in image and data analysis to learn more intricate patterns as well as to generate novel samples. Variational autoencoders are appreciated for ensuring a clear format and stable training and are used in recovery, learning features, and data augmentation. One of the frequent drawbacks is that pixel-based loss will result in output with missing details and blurred images. To better this, researchers have experimented with improvements to the latent space, conditional generation mechanisms, and feature-based perceptual losses, which assist in maintaining coherence of structure as well as maintaining Variational Autoencoder (VAE) stability. This paper examines the history of VAEs, their main technical advance, and their use in reconstruction, representation learning, and data generation. Meanwhile, this paper also points out the main challenges currently faced, such as insufficient retention of details, high computational costs, and limited cross-domain adaptability, and looks forward to possible future research directions.

Copyright
© 2026 The Author(s)
Open Access
Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

<Previous Article In Volume
Next Article In Volume>
Volume Title
Proceedings of the International Workshop on Advances in Deep Learning for Image Analysis and Computer Vision (IWADIC 2025)
Series
Advances in Computer Science Research
Publication Date
24 April 2026
ISBN
978-94-6239-648-7
ISSN
2352-538X
DOI
10.2991/978-94-6239-648-7_96How to use a DOI?
Copyright
© 2026 The Author(s)
Open Access
Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Cite this article

risenwbib
TY  - CONF
AU  - Xuan Sheng
PY  - 2026
DA  - 2026/04/24
TI  - Research and Analysis of VAE in Image and Data Analysis
BT  - Proceedings of the International Workshop on Advances in Deep Learning for Image Analysis and Computer Vision (IWADIC 2025)
PB  - Atlantis Press
SP  - 893
EP  - 902
SN  - 2352-538X
UR  - https://doi.org/10.2991/978-94-6239-648-7_96
DO  - 10.2991/978-94-6239-648-7_96
ID  - Sheng2026
ER  -