DeepFakes explained + Hottest girls on the net Part 2

Internet and deepfake are not just for porn but also serve as an excellent tool to make a reliable survey

In the first part of this post I explained the algorithm behind deepfake that is able to create highly realistic face swap on videos.

But many of you asked “Okay, but where are the hot girls?”. This second part will answer this question.

How to find the hottest girls?

It is a common problem in surveys that what people say is not always what they actually believe. Not because in any way they would intend to misguide us but simply because it is not always easy to quickly answer even a question that looks so simple like that. There are so many nice and famous girls out there how could we select in some seconds the best ones. Probably we can quickly select nice ones but I am not sure they would be really the most stunning ones.

Sometimes it is difficult to choose

A good workaround for this problem if people have to make some effort before they can select their preferred solution. The bigger the effort the higher the chance that people will make very thorough and careful evaluation before they make their final decisions. In these cases even very small sample sizes can lead to better results than big surveys. A good example for this are the betting markets that are much better in predicting the outcome of certain events than surveys conducted on a much bigger sample. Here effort is represented by the money that we use to bet on certain outcomes. Certainly we are much more careful in making a decision when our money is on stake.

Not only money but also time can be considered as a kind of effort. And do you know what takes a lot of time? Creating deepfake videos. It can easily take 10–12 hours even on a computer with a decent GPU to produce one. Certainly you will think long what is really interesting for you or for your target audience before starting to create one. If we can somehow collect this data we will have a very good survey.

My idea was to collect from the deepfakes subreddit all the post titles. From these titles we can extract the name of deepfaked persons, count them and create our ranking.

Show me the code

Let’s see how can we do that in Python in one of the greatest (at least according to me 😄) all around programming language.

import praw
import pandas as pd
import re

In Python we have a nice package for almost every tasks. We start with importing them:

• praw: to interact with Reddit
• pandas: to have a convenient data format
• re: regular expression to extract names

reddit = praw.Reddit(client_id=’my_id’,
 client_secret=’my_secret’,
 user_agent=’my user agent’)
titles=[]
for submission in reddit.subreddit('deepfakes').hot(limit=None):
    titles.append(submission.title)
name_pattern=re.compile("(?=([A-Z][a-z]+\s[A-Z][a-z]+))")
names=[name for title in titles for name in name_pattern.findall(title)]
print("{} name like strings ({} unique) out of {} titles extracted.".format(len(names),len(set(names)), len(titles)))

581 name like strings (443 unique) out of 988 titles extracted.

Using praw we downloaded all post titles from the deepfakes subreddit. We assume that a name consist of two words where the first character of each word is capitalized and the rest is lowercase. We extract names with that pattern from the titles.

The net is crazy about Nicolas Cage but he is very often just referenced as Nic Cage. Let’s merge this two names since they represent the same person.

nic_cage_count=sum([1 if (name==”Nic Cage”) else 0 for name in names])
nicolas_cage_count=sum([1 if (name==”Nicolas Cage”) else 0 for name in names])
names=[“Nic Cage/Nicolas Cage” if (name==”Nic Cage” or name==”Nicolas Cage”) else name for name in names]
print(“{} Nic Cage and {} Nicolas Cage found. {} altogether.”.format(nic_cage_count,nicolas_cage_count,sum([nic_cage_count,nicolas_cage_count])))

9 Nic Cage and 5 Nicolas Cage found. 14 altogether.

names=pd.DataFrame(data=names, columns=[“name”])
names=pd.DataFrame(names[“name”].value_counts().reset_index())
names.columns=[“name”,”count”]
names

We use Pandas this awesome Python library to do data manipulation effectively. Certainly we can do it just using simple array operations but believe me it would be much longer and less elegant. (I actually did that first so you can really trust me on that…) In just one line we can count the names and order them.

Finally the hot girls

But now lets see the results (names and counts) that we all waited for so long.

1. Emma Watson (15)
2. Nic Cage/Nicolas Cage (14)
3. Daisy Ridley (12)
4. Natalie Dormer (11)
5. Danielle Panabaker (6)
6. Scarlett Johansson, Grace Moretz, Gal Gadot (5)
7. Ariana Grande, Ivanka Trump, Emilia Clarke, Cara Delevingne, Taylor Swift, Natalie Portman, Maisie Williams, Sophie Turner (4)
8. Elsa Jean, Jessica Chastain, Donald Trump, Michelle Obama (3)

Wow, Emma Watson is the hottest girl on the net according to the deepfake videos, big applause for her!

But everyone's all time favorite Nic Cage is a really strong contender. Probably he is preferred for other skills than Emma.

Well deserved Nic!

Interesting to see that many actress playing super hero roles and from Game of Thrones got a high position.

The Trump family has two members in the top 8. In case of Ivanka it is probably more obvious than for Donald. Michelle Obama is the only other politician who could reach the top league.

Interestingly we have also a porn star here Elsa Jean. Probably her face was not inserted into the movies rather the other way around. (though it would be an interesting approach as well)

Wow, in just two short posts we’ve understood how deepfake works and we have also seen an interesting solution how to utilize this data for a cheap and effective survey. (certainly not forget about the hot girls!) Good job, thanks for joining me on this journey!

Please clap 👏 👏 👏 if you enjoyed and follow me for more interesting posts where we have fun with the data out there and explore new ways to analyze them!