Real-Time Detector Sees Hand Poses and Tracks Multiple People
Researchers at Carnegie Mellon University’s Robotics Institute have enabled a computer to understand the body poses and movements of multiple people from video in real time — including, for the first time, the pose of each individual’s fingers.
This new method was developed with the help of the Panoptic Studio, a two-story dome embedded with 500 video cameras. The insights gained from experiments in that facility now make it possible to detect the pose of a group of people using a single camera and a laptop computer.
Yaser Sheikh, associate professor of robotics, said these methods for tracking 2-D human form and motion open up new ways for people and machines to interact, and for people to use machines to better understand the world around them. The ability to recognize hand poses, for instance, will make it possible for people to interact with computers in new and more natural ways, such as communicating with computers simply by pointing at things.
Detecting the nuances of nonverbal communication between individuals will allow robots to serve in social spaces, allowing robots to perceive what people around them are doing, what moods they are in and whether they can be interrupted. A self-driving car could get an early warning that a pedestrian is about to step into the street by monitoring body language. Enabling machines to understand human behavior also could lead to new approaches to behavioral diagnosis and rehabilitation for conditions such as autism, dyslexia and depression.
“We communicate almost as much with the movement of our bodies as we do with our voice,” Sheikh said. “But computers are more or less blind to it.”
In sports analytics, real-time pose detection will make it possible for computers to not only track the position of each player on the field of play, as is now the case, but also to know what players are doing with their arms, legs and heads at each point in time. The methods can be used for live events or applied to existing videos.
To encourage more research and applications, the researchers have released their computer code for both multiperson and hand-pose estimation. It already is being widely used by research groups, and more than 20 commercial groups, including automotive companies, have expressed interest in licensing the technology, Sheikh said.
Sheikh and his colleagues will present reports on their multiperson and hand-pose detection methods at CVPR 2017, the Computer Vision and Pattern Recognition Conference, July 21–26 in Honolulu.
Tracking multiple people in real time, particularly in social situations where they may be in contact with each other, presents a number of challenges. Simply using programs that track the pose of an individual does not work well when applied to each individual in a group, particularly when that group gets large. Sheikh and his colleagues took a bottom-up approach, which first localizes all the body parts in a scene — arms, legs, faces, etc. — and then associates those parts with particular individuals.
The challenges for hand detection are even greater. As people use their hands to hold objects and make gestures, a camera is unlikely to see all parts of the hand at the same time. Unlike the face and body, large datasets do not exist of hand images that have been laboriously annotated with labels of parts and positions.
But for every image that shows only part of the hand, there often exists another image from a different angle with a full or complementary view of the hand, said Hanbyul Joo, a Ph.D. student in robotics. That’s where the researchers made use of CMU’s multicamera Panoptic Studio.
“A single shot gives you 500 views of a person’s hand, plus it automatically annotates the hand position,” Joo explained. “Hands are too small to be annotated by most of our cameras, however, so for this study we used just 31 high-definition cameras, but were still able to build a massive data set.”
Joo and Tomas Simon, another Ph.D. student, used their hands to generate thousands of views.
“The Panoptic Studio supercharges our research,” Sheikh said. It now is being used to improve body, face and hand detectors by jointly training them. Also, as work progresses to move from the 2-D models of humans to 3-D models, the facility’s ability to automatically generate annotated images will be crucial.
When the Panoptic Studio was built a decade ago with support from the National Science Foundation, it was not clear what impact it would have, Sheikh said.
“Now, we’re able to break through a number of technical barriers primarily as a result of that NSF grant 10 years ago,” he added. “We’re sharing the code, but we’re also sharing all the data captured in the Panoptic Studio.”
In addition to Sheikh, the multiperson pose-estimation research included Simon and master’s degree students Zhe Cao and Shih-En Wei. The hand-detection study included Sheikh, Joo, Simon and Iain Matthews, an adjunct faculty member in the Robotics Institute. Gines Hidalgo Martinez, a master’s degree student, also collaborates on this work, managing the source code.
Learn more: A Computer That Reads Body Language
The Latest on: A computer that can read body language
- KIM PURSCELL: Qualities of a modern gentleman never go out of style on October 6, 2018 at 10:02 am
His intellectual curiosity propels him to constantly better himself and his craft through reading, coaching ... He manages his tone and body language to maximize connection. He knows that hearing is a ... […]
- 12 job interview tricks that are hard to master, but will pay off forever on October 6, 2018 at 5:00 am
The hiring manager wants to hear more about the experiences he read about on paper, and I promise you no one is looking to see how much shaking you can do in those boots ... Maintain good body languag... […]
- What we lose by reading 100,000 words every day on October 5, 2018 at 7:41 am
She compares the many elements that reading sets in motion — vision, language ... Digital technology can perpetuate inequalities as well as solve them. Not every kid grows up with books in the house; ... […]
- How To 'Read' Subconscious Body Language And Attraction on October 3, 2018 at 8:30 am
Nowhere is this more true than in the dating world - where an errant sideways look can signal everything ... with a in-depth understanding of human body language. These 'translation' tips will help yo... […]
- These Bots Have Eyes: Why the Evolution of Visual Chatbots Is a Boon for Entrepreneurs on September 24, 2018 at 9:30 am
These are emotional, visual creatures that communicate with body language ... arrived. Computer vision AI is already being utilized in a wide range of applications. It recognizes faces and smiles in c... […]
- WATCH: Rep. Paul Gosar Says He Reads Body Language Well Because He’s a Dentist on July 12, 2018 at 4:47 pm
So I read body language very, very well ... Strzok told Gowdy during that exchange, according to CBS News, “I can assure you Mr. Chairman, at no time in any of these texts did those personal beliefs e... […]
- CMU researchers create a huge dome that can read body language on July 7, 2017 at 1:12 am
The Panoptic Studio is a new body scanner created by researchers at Carnegie Mellon University that will be used to understand body language in real situations ... a group haggling over an object. The ... […]
- How to Ace a Job Interview on February 17, 2015 at 6:29 am
"Phone interviews are much more difficult," Hyatt says. "Sometimes the connection doesn't work. You can't read body language, and you feel like you're [talking over] people." Before the interview star... […]
- The Confidence Game: How to Bring Out Your Inner Hotshot on November 14, 2012 at 5:25 am
So when a friend tells me about an executive education class for women called "Acting with Power," I can't help seeing it as a sign ... an expert in the body language of power. (Today's lesson is an a... […]
via Google News and Bing News