Summary of A Survey Of Synthetic Data Augmentation Methods in Computer Vision, by Alhassan Mumuni et al.
A survey of synthetic data augmentation methods in computer vision
by Alhassan Mumuni, Fuseini Mumuni, Nana Kobina Gerrar
First submitted to arxiv on: 15 Mar 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Graphics (cs.GR); Machine Learning (cs.LG)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The paper presents a comprehensive review of synthetic data augmentation techniques for computer vision problems. Specifically, it explores methods based on realistic 3D graphics modeling, neural style transfer (NST), differential neural rendering, and generative artificial intelligence (AI) techniques such as generative adversarial networks (GANs) and variational autoencoders (VAEs). The paper covers data synthesis approaches, highlighting important generation and augmentation techniques, scope of application, specific use-cases, existing limitations, and possible workarounds. Additionally, it provides an overview of common synthetic datasets for training computer vision models, emphasizing main features, application domains, and supported tasks. Finally, the paper discusses the effectiveness of synthetic data augmentation methods. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Synthetic data augmentation is a way to overcome challenges in obtaining sufficient image data for target tasks in computer vision. The paper reviews various techniques like 3D graphics modeling, neural style transfer, differential neural rendering, and GANs/VAEs. These methods help generate training data from scratch, which can be used to train deep convolutional neural networks (CNNs). The review covers the benefits and limitations of each method, as well as common synthetic datasets for computer vision tasks. |
Keywords
* Artificial intelligence * Style transfer * Synthetic data
