Short answer

Do not solely rely on popularity metrics; explore objective quality assessment tools to find exceptional visual content.

Field: Classic Design
Source: Institutional Research Information System University of Turin (University of Turin) (2015)
Method: Computer Vision and Crowdsourced Evaluation
Sample: Large dataset of creative-commons photos on Flickr with crowdsourced aesthetic scores
Evidence: Strong effect

Automated beauty assessment can identify aesthetically superior content that has low social media visibility. This classic design research insight is drawn from a 2015 study published in Institutional Research Information System University of Turin (University of Turin). Using Computer vision and crowdsourced evaluation with Large dataset of creative-commons photos on Flickr with crowdsourced aesthetic scores, researchers explored how this design variable affects real-world outcomes. The key design takeaway: Do not solely rely on popularity metrics; explore objective quality assessment tools to find exceptional visual content.

Study

Classic DesignHigh ImpactStrong effect

Objective beauty metrics can surface overlooked high-quality visual content

Automated beauty assessment can identify aesthetically superior content that has low social media visibility.

Institutional Research Information System University of Turin (University of Turin) · 2015

Key Findings

01The computer vision method successfully identified photos with high aesthetic quality.
02The beauty scores of photos surfaced by the method were comparable to those of the most popular photos.
03The average beauty score of surfaced photos was only slightly lower than the most popular ones, indicating a significant discovery of high-quality, low-visibility content.

Application

Design takeaway

Do not solely rely on popularity metrics; explore objective quality assessment tools to find exceptional visual content.

How to apply

Use image analysis software or AI tools that offer aesthetic scoring to filter through large collections of images for design projects, seeking out visually compelling but less common examples.

Project actions

01Consider how to objectively measure the quality of visual elements in your design.
02Explore tools that can analyze visual aesthetics programmatically.

Method & Evidence

AimCan objective beauty metrics be used to identify visually high-quality content that is currently overlooked due to low popularity on social media platforms?

MethodComputer Vision and Crowdsourced Evaluation

ProcedureA computer vision model was developed to assess the aesthetic quality of images. This model's performance was then evaluated against a large dataset of Flickr photos, for which aesthetic scores were collected through crowdsourcing. The model's ability to identify 'beautiful' images was compared to their popularity metrics.

SampleLarge dataset of creative-commons photos on Flickr with crowdsourced aesthetic scores

ContextSocial media photo-sharing platforms (e.g., Flickr)

Variables

IVImage aesthetic quality (as determined by computer vision model)

DVPerceived beauty score (from crowdsourcing) and popularity metrics (e.g., favorites, views)

CVImage content, resolution, source platform (Flickr)

Strengths & Limitations

Strengths

+Utilizes a large-scale dataset.
+Combines computational analysis with human perception data.

Limitations

The subjectivity of beauty and the potential for AI bias in aesthetic judgment.

Reliability & validity

Reliability of crowdsourced scores can be improved by averaging many ratings. Validity is supported by comparing algorithmic scores to human consensus.

Think critically

To what extent can an algorithm truly capture the nuanced and subjective nature of aesthetic beauty, and what are the ethical implications of relying on such algorithms for content curation?

Design Principles

"Prioritize intrinsic quality over perceived popularity when evaluating visual assets."

In design practice, especially in fields involving visual content like graphic design, product visualization, or digital art, a significant amount of high-quality work can go unnoticed due to social media algorithms prioritizing popularity over intrinsic merit. This research suggests that objective beauty metrics can act as a valuable tool to discover and promote such hidden gems, ensuring that quality is not overshadowed by sheer visibility.

What This Means for Your Design

Computers can learn to spot beautiful pictures, even if they aren't famous online, helping designers find great images that might otherwise be missed.

How to use in your project

1.This research can inform the selection of visual references or the evaluation of design outcomes by suggesting objective quality measures.

Add to My Project

Quick Cite

Paragraph starter

This study highlights that objective beauty metrics, as demonstrated by computer vision models, can effectively surface high-quality visual content that may be overlooked due to low popularity on social media. This suggests that designers can employ similar analytical approaches to identify aesthetically superior assets, ensuring that intrinsic quality is not overshadowed by visibility, thereby enriching the inspiration pool for design projects.

Source

Institutional Research Information System University of Turin (University of Turin)

An Image Is Worth More than a Thousand Favorites: Surfacing the Hidden\nBeauty of Flickr Pictures

journal · 2015

View source

Questions About This Research

What does the research say about objective beauty metrics can surface overlooked high-quality visual content?: Do not solely rely on popularity metrics; explore objective quality assessment tools to find exceptional visual content. Evidence: Institutional Research Information System University of Turin (University of Turin) (2015).
Why does "Objective beauty metrics can surface overlooked high-quality visual content" matter for design?: In design practice, especially in fields involving visual content like graphic design, product visualization, or digital art, a significant amount of high-quality work can go unnoticed due to social media algorithms prioritizing popularity over intrinsic merit. This research suggests that objective beauty metrics can act as a valuable tool to discover and promote such hidden gems, ensuring that quality is not overshadowed by sheer visibility.
How can designers apply this research?: Do not solely rely on popularity metrics; explore objective quality assessment tools to find exceptional visual content.
What were the main findings?: The computer vision method successfully identified photos with high aesthetic quality.. The beauty scores of photos surfaced by the method were comparable to those of the most popular photos.. The average beauty score of surfaced photos was only slightly lower than the most popular ones, indicating a significant discovery of high-quality, low-visibility content.
What research method was used?: Computer Vision and Crowdsourced Evaluation with Large dataset of creative-commons photos on Flickr with crowdsourced aesthetic scores.
How strong is the evidence?: Evidence strength is rated Strong effect, based on a 2015 journal from Institutional Research Information System University of Turin (University of Turin).
What should I do differently in my next project?: Use image analysis software or AI tools that offer aesthetic scoring to filter through large collections of images for design projects, seeking out visually compelling but less common examples.
What are the limitations?: The definition of 'beauty' can be subjective and culturally influenced; the computer vision model's aesthetic judgment may not align with all human perceptions.