Image Analysis through the lens of ChatGPT-4

https://doi.org/10.48185/jaai.v4i2.870

Authors

  • Olanrewaju Victor Johnson Federal Polytechnic Ile-Oluji, Ondo State, PMB 727, Nigeria
  • Osamah Mohammed Alyasiri Karbala Technical Institute, Al-Furat Al-Awsat Technical University, 56001, Karbala
  • Dua’a Akhtom School of computer sciences, University Sains Malaysia, Penang 11800, Malaysia
  • Olabisi Esher Johnson Federal Polytechnic Ile-Oluji, Ondo State, PMB 727, Nigeria

Keywords:

Artificial Intelligence, ChatGPT, Images Analysis, Image Content Extraction

Abstract

Numerous studies have delved into the applications of ChatGPT across various domains such as medicine, sports, education, and business analysis. ChatGPT emerges as a potential replacement for key contributors in these diverse fields, sparking an ongoing quest to validate this assertion. One focal point of this paper is the examination of GPT-4's,  the fourth generation of Chat GPT, capacity to handle a spectrum of visual elements like images, pictures, flowcharts, plots, and diagrams. The inquiry extends to assessing how the gleaned information from these visuals compares with human intuition, both inductive and deductive. To investigate, GPT-4 was presented with samples of human faces, flowcharts, plots, and diagrams, leading to remarkably accurate and error-free results within the specified timeframe, surpassing human capabilities. The outcomes underscore GPT-4's impressive prowess in image analysis, covering identification, recognition, and contextual understanding of visual content. Furthermore, GPT-4's proficiency in identifying objects within individual  images opens the door to be utilized comprehensively in the field of object detection. However, GPT-4 exhibits limitations in recognizing individual images due to privacy considerations.

Downloads

Download data is not yet available.

References

P. Handa, D. Chhabra, N. Goel, and S. Krishnan, “Exploring the role of ChatGPT in medical image analysis,” Biomed Signal Process Control, vol. 86, no. PC, p. 105292, 2023, doi: 10.1016/j.bspc.2023.105292.

J. S. Tech, “Exploring the Landscape of Modern Machine Learning and AI.” Accessed: Oct. 17, 2023. [Online]. Available: https://jillanisofttech.medium.com/exploring-the-landscape-of-modern-machine-learning-and-ai-cf57d8ff9637

J. O. Victor, X. Chew, K. W. Khaw, and M. H. Lee, “A Cost-Based Dual ConvNet-Attention Transfer Learning Model for ECG Heartbeat Classification,” Journal of Informatics and Web Engineering, vol. 2, no. 2, pp. 90–110, 2023, doi: 10.33093/jiwe.2023.2.2.7.

O. Diaz-Hernandez and V. J. Gonzalez-Villela, “Analysis of Human Intuition Towards Artificial Intuition Synthesis for Robotics,” SSRN Electronic Journal, vol. 1, no. 1, pp. 23–39, 2019, doi: 10.2139/ssrn.3427647.

Published

2023-12-28

How to Cite

Johnson, O. V. ., Mohammed Alyasiri, O., Akhtom, D. ., & Johnson, O. E. . (2023). Image Analysis through the lens of ChatGPT-4. Journal of Applied Artificial Intelligence, 4(2), 31–46. https://doi.org/10.48185/jaai.v4i2.870

Issue

Section

Articles

Most read articles by the same author(s)