Over a million developers have joined DZone.
{{announcement.body}}
{{announcement.title}}

Placemeter: Analytics for the Real World

DZone's Guide to

Placemeter: Analytics for the Real World

We can track online transactions with a few lines of code, but can the same be made true for real-world premises? With Placemeter, this might be possible

· IoT Zone
Free Resource

Download Red Hat’s blueprint for building an open IoT platform—open source from cloud to gateways to devices.

Every day we're inundated with a new app for this, a new site for that or a new device for something else. Like most of us, I use ProductHunt to keep an eye on the most valued of these products. Placemeter caught my attention a few weeks back, with it's claim that it "turns video into meaningful data". With this combination of computer vision and data analytics,  I had to find out more. 

Before you go any further, check out this overview of what Placemeter can do 

The People Behind Placemeter 

First a little on the background of the company and it's founders. Placemeter was co-founded in 2012 by Alexandre Winter and Florent Peyre. Alex is a computer vision expert and a serial entrepreneur. Prior to Placemeter, Alex co-founded LTU Technologies, a venture-backed startup that was a pioneer in large-scale image search by content. As the CTO—and later, the CEO—Alex helped bring LTU to profitability in three years. LTU was acquired by a Japanese company listed on the Tokyo Stock Exchange in 2005. Prior to LTU, Alex was a researcher in Computer Vision at INRIA. He also worked at Airbus building vision-based missile guidance systems.

Alex holds an MSc and a PhD in Image Recognition from Telecom ParisTech. He also holds nine patents in the field of computer vision. In his free time, Alex advises various startups in New York and is a mentor of the Techstars NYC program.

Florent Peyre is the Co-Founder & COO of Placemeter. Previously, Florent started Casahop, a social travel company in the home exchange space. He founded the company with Paul Berry, the ex-CTO of the Huffington Post and now CEO of Rebelmouse. The company was backed by Lerer Ventures, First Round Capital, and other prominent investors.

Prior to Casahop, Florent served as the Vice President of Strategy & Business Development for Gilt City, member of the Gilt Groupe, and was one of its founding members. He scaled the business to 150 employees and several million dollars in revenue per month. Florent also ran national sales for the company.

Before Gilt City, Florent presided over Hachette Filipacchi Media’s Emerging Platforms group and launched several mobile and iPad applications for Hachette’s brands. There he also signed several strategic deals including the launch of a new lifestyle destination for Microsoft (Glo).

Florent holds an MA in International Law from Sorbonne University and a Master’s degree from Sciences Po Paris. In his free time he serves as an advisor to several New York startups and as a mentor to SeedStart NY and First Growth Venture Networks.

Obviously both Alex and Florent have the previous experience to make something like Placemeter really work.

The Genesis a Great Idea

Placemeter was born out of the desire to help people and businesses better understand and use the world around them.

After moving to New York City, the first thing Alex’s kids wanted to do was go to Shake Shack. They waited in line for over an hour. While waiting, Alex learned of the publicly available Shack Cam, which people can access to avoid that notoriously long line. Leveraging his background in computer vision, Alex went home to analyze how the line moved. He found that the line didn’t move at the same speed all the time – it depended on the number of counters open, the length of the line, and several other variables. With new insight, Alex found it was possible to predict the best time to go to Shake Shack to get the most enjoyable experience.

In his first months in New York, Alex quickly realized that density was a growing problem with no solution in sight. He also saw an information imbalance—the data that describes how cities move and grow was in hands of few rather than many.

Placemeter was conceived to make urban data accessible to anyone who wants to improve their community.

It Goes Beyond Retail

One of the first assumptions I made when I saw the introduction video what that this was all about the retail industry. However, I was corrected about this misconception

Placemeter is an open platform that can be used by everyone, for various use cases and in various industries. That is an important differentiator for us; our competitors primarily focus on only one vertical. We believe our technology has more uses beyond a single application.  That said, retailers are primary customers, as are cities and their governments. - Alexandre Winter, CEO and co-founder, Placemeter

Retailers use Placemeter to better understand foot traffic around their stores. The data helps them reveal hidden patterns and uncover strategic opportunities. With a Placemeter Sensor, retailers can select the most profitable sites for new stores, optimize hours and staffing for maximum engagement, and analyze the effect of marketing and merchandising campaigns.

Cities and local governments can leverage existing infrastructure to understand how people use public spaces. They can use the data to optimize traffic and pedestrian flows at busy intersections, sell advertising space on bus shelters and in parks, and provide citizens with before and after measurements of capital improvement projects.

Along with local governments and businesses, Placemeter supports non-revenue generating groups including civic hackers, nonprofits, academics, and researchers. The company’s mission is to build better cities, and these groups help achieve those goals in unique ways.

What Can The Camera Detect?

The burning question for me was whether it could detect the difference between the person that it was looking at; whether it was a pedestrian, a cyclist or a motorist. 

Today we track pedestrian, bicycle, and vehicle traffic the same way. Users can break down the measurements by setting up separate measurement points for sidewalks, bike lanes, and roads.

However, we know that in many areas, different types of traffic use the same paths, and our users may be interested in the mix. It’s reasonable to expect that we’ll release a new and improved algorithm in early 2016 that can differentiate between cars, buses, motorcycles, trucks, vans, ambulances, bicycles and more. To do so, we have to train the computers to recognize the different ways the objects look and move. We’re actively building toward that goal, and we’re almost there.
- Alexandre Winter, CEO and co-founder, Placemeter


So How Does A Developer Use This?

The API enables users to pull hourly Traffic Counts (the number of vehicles, bicycles, or pedestrians that pass a designated measurement point every hour), Directional Counts (the number of people that cross the measurement point in each direction every hour), and Occupancy Counts (the number of people within an area, every hour, and the number of people come in and out of that area).


Our customers are gathering a ton of data. Our API helps them synthesize and aggregate that data. They can fold the data into a mapping application, like CartoDB or ArcGIS, to discover patterns and trends in movement. They can correlate our data with their proprietary data, such as POS or sales data, to answer specific questions. They can also bake our data right into their own dashboards. At the highest level, we quantify how cities move, and an API is the perfect way to consume that data. - Alexandre Winter, CEO and co-founder, Placemeter

I assumed there would be some element of iBeacon, or Near Field Technology in effect with the product but found that Placemeter place more emphasis on computer vision 

Computer vision allows us to be simultaneously very accurate and very scalable. And with computer vision you can use one sensor to measure many things, whereas beacons are typically purpose-built to answer one question.

We see another advantage in the long-term. Our technology will develop over time, and upgrading your sensor to the latest version will be as simple as updating an app. Other solutions require full swap-outs.
- Alexandre Winter, CEO and co-founder, Placemeter


Placemeter have  just announced a new API and a new Sensor. And, in 2016, they are planning to develop a more robust algorithm that will give our customers a new layer of precision, along with a  plan to build more powerful analytics to sit on top of the urban intelligence platform they've developed today. In the next few weeks I'll be taking a deeper dive and trying out the product for myself. 

Build an open IoT platform with Red Hat—keep it flexible with open source software.

Topics:
data ,analytics ,sensor

Opinions expressed by DZone contributors are their own.

{{ parent.title || parent.header.title}}

{{ parent.tldr }}

{{ parent.urlSource.name }}