Summary of Perception Of Visual Content: Differences Between Humans and Foundation Models, by Nardiena A. Pratama et al.
Perception of Visual Content: Differences Between Humans and Foundation Models
by Nardiena A. Pratama, Shaoyang Fan, Gianluca Demartini
First submitted to arxiv on: 28 Nov 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: 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 compares human-generated and ML-generated annotations of images representing diverse socio-economic contexts, aiming to understand differences in perception and identify potential biases. The dataset consists of images of people washing their hands from various geographical regions and income levels. Semantically, the results show low similarity between human and machine annotations at a low-level perspective, but they are alike in how they perceive images across different regions. Additionally, human annotations resulted in better overall and more balanced region classification performance on the class level, while ML Objects and ML Captions performed best for income regression. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The study compares how humans and machines label pictures of people from different places and with different incomes. The pictures are of people washing their hands, which helps understand how people from different backgrounds might be perceived or judged differently. Humans and machines have similar ideas about what the images mean, but they use different words to describe them. When it comes to classifying where the people come from, human labels are better at getting the right answer most of the time. But for predicting income levels, machine-generated labels work best. |
Keywords
» Artificial intelligence » Classification » Regression
