Human Pose Estimation: The Tech That Enables Multiple AI Applications
In our previous blog posts we have discussed how AI has been used in the Olympics and by Yoga practitioners. It gets one curious about the technology that enables such advanced facilitations. Human Pose Estimation, an application of Computer Vision is what enables these functions in a near accurate manner. PoseNet is a deep learning TensorFlow open-source model that enables one to figure as well as track human poses by spotting body parts like arms, knees, elbows, hips, wrists etc in a photograph or a video. A Human Pose Skeleton is essentially a set of coordinates that describe an individual’s pose while a limb is a pair of these coordinates. The human skeleton is numbered in points and the pose is estimated by identification and location of these numbers in the video or the photograph.
Human pose estimation can be used for a number of uses including gesture recognition, animation and gaming, visual search, in the training of robots, in a computer-aided diagnosis, for building augmented reality experiences, and for activity recognition for real-time sports analysis such as Olympics.
This Olympics may have been closed to many due to the pandemic, but all minute details of the Tokyo games could be captured with the help of a 3-D tracking system. If the advanced tech used at the Olympics is anything to go by, it also suggests that in the future there will be digital replicas of the human body at the games which will be assisting the athletes in getting better at their sport. AI showed that in the NBC broadcast of the 100 meter trials in Eugene, Ore., Sha’Carri Richardson hit 24.1 miles per hour when she was at her peak, but she slowed to hit 20.0 mph near the finish line but she still won. The runner-up, however, peaked at 23.2 miles per hour, slowing down at 20.4 mph near the finish line. When the runner up analyses these results it can help improve the race enough to even surpass Richardson and that explains the crucial role of AI and human pose estimation at Olympics.
Director of sports performance technology in the Olympic technology group at Intel, Jonathan Lee, used a lot of footage from the field and the track while they were in motion to train the Olympic AI using ML. The movements of the human body were recorded while a team manually explained each part of the body down to details such as pixels. After this, the AI model was able to track the skeleton and performed a 3-D pose estimation of the athletes during the event. By the next Olympics similar tech will be used across other sports too. In fact non-athletes such as yoga practitioners and those going to the gym could also benefit from this tech.
It’s most obvious in the digital media space, from click buys to personalized web experiences. For marketing, the AI journey has just kick-started, while in the tech sector it has been applied for a while now. We are still at an early stage where inroads are being made into AI content via chatbots and even some explanatory content creation but what will make anyone jump up and embrace it is when we will start seeing a lot of mainstream content being created by AI.
Prior to joining Infinite Analytics, Richard served as the CFO of CrowdFlower, COO and CFO of Phoenix Technologies, as a member of the board of directors and chairman of the Audit Committee at Intellisync, and previously as CFO and executive vice president strategy and corporate development at Charles Schwab.
Pravin Gandhi has over 50 years of entrepreneurial operational and investing experience in the IT industry in India. He was a founding partner of the first early stage fund India - INFINITY. Subsequently a founding partner in Seedfund I & II. With over 18 years of investing experience, he is extensively well networked in investment and entrepreneurial scene and is an active early stage angel investor in tech & impact space. Pravin holds a BS in Industrial Engineering from Cornell University, and serves on the board of several private corporations in India. He is on the board of SINE, IIT Mumbai Incubator.
Puru has his Masters in Engineering and Management from MIT. Prior to MIT, he worked with Fidelity Investments building electronic trading products and high volume market data processing applications. He has completed his BE from VJTI, Mumbai.
Deb Roy is Professor of Media Arts and Sciences at MIT where he directs the MIT Center for Constructive Communication, and a Visiting Professor at Harvard Law School. He leads research in applied machine learning and human-machine interaction with applications in designing systems for learning and constructive dialogue, and for mapping and analyzing large scale media ecosystems. Deb is also co-founder and Chair of Cortico, a nonprofit social technology company that develops and operates the Local Voices Network to surface underheard voices and bridge divides.
Roy served as Executive Director of the MIT Media Lab from 2019-2021. He was co-founder and CEO of Bluefin Labs, a media analytics company that analyzed the interactions between television and social media at scale. Bluefin was acquired by Twitter in 2013, Twitter’s largest acquisition of the time. From 2013-2017 Roy served as Twitter’s Chief Media Scientist.
Erik Brynjolfsson is the Jerry Yang and Akiko Yamazaki Professor and Senior Fellow at the Stanford Institute for Human-Centered AI (HAI), and Director of the Stanford Digital Economy Lab. He also is the Ralph Landau Senior Fellow at the Stanford Institute for Economic Policy Research (SIEPR), Professor by Courtesy at the Stanford Graduate School of Business and Stanford Department of Economics, and a Research Associate at the National Bureau of Economic Research (NBER).
Akash co-founded IA while studying for his MBA from MIT. Prior to MIT Sloan, he co-founded Zoonga. Before this, Akash was an engineer with Oracle in Silicon Valley. He has completed his M.S from University of Cincinnati and B.E from the College of Engineering, Pune.