Egocentric vision explained

Egocentric vision or first-person vision is a sub-field of computer vision that entails analyzing images and videos captured by a wearable camera, which is typically worn on the head or on the chest and naturally approximates the visual field of the camera wearer. Consequently, visual data capture the part of the scene on which the user focuses to carry out the task at hand and offer a valuable perspective to understand the user's activities and their context in a naturalistic setting.^[1]

The wearable camera looking forwards is often supplemented with a camera looking inward at the user's eye and able to measure a user's eye gaze, which is useful to reveal attention and to better understand theuser's activity and intentions.

History

The idea of using a wearable camera to gather visual data from a first-person perspective dates back to the 70s, when Steve Mann invented "Digital Eye Glass", a device that, when worn, causes the human eye itself to effectively become both an electronic camera and a television display.^[2]

Subsequently, wearable cameras were used for health-related applications in the context of Humanistic Intelligence^[3] and Wearable AI.^[4] Egocentric vision is best done from the point-of-eye, but may also be done by way of a neck-worn camera when eyeglasses would be in-the-way.^[5] This neck-worn variant was popularized by way of the Microsoft SenseCam in 2006 for experimental health research works.^[6] The interest of the computer vision community into the egocentric paradigm has been arising slowly entering the 2010s and it is rapidly growing in recent years,^[7] boosted by both the impressive advances in the field of wearable technology and by the increasing number of potential applications.

The prototypical first-person vision system described by Kanade and Hebert,^[8] in 2012 is composed by three basic components: a localization component able to estimate the surrounding, a recognition component able to identify object and people, and an activity recognition component, able to provide information about the current activity of the user. Together, these three components provide a complete situational awareness of the user, which in turn can be used to provide assistance to the user or to the caregiver. Following this idea, the first computational techniques for egocentric analysis focused on hand-related activity recognition ^[9] and social interaction analysis.^[10] Also, given the unconstrained nature of the video and the huge amount of data generated, temporal segmentation^[11] and summarization^[12] were among the first problems addressed. After almost ten years of egocentric vision (2007 - 2017), the field is still undergoing diversification. Emerging research topics include:

• Social saliency estimation^[13]
• Multi-agent egocentric vision systems
• Privacy preserving techniques and applications
• Attention-based activity analysis^[14]
• Social interaction analysis^[10]
• Hand pose analysis^[15]
• Ego graphical User Interfaces (EUI)^[16]
• Understanding social dynamics and attention^[17]
• Revisiting robotic vision and machine vision as egocentric sensing ^[18]
• Activity forecasting^[19]

Technical challenges

Today's wearable cameras are small and lightweight digital recording devices that can acquire images and videos automatically, without the user intervention, with different resolutions and frame rates, and from a first-person point of view. Therefore, wearable cameras are naturally primed to gather visual information from our everyday interactions since they offer an intimate perspective of the visual field of the camera wearer.

Depending on the frame rate, it is common to distinguish between photo-cameras (also called lifelogging cameras) and video-cameras.

• The former (e.g., Narrative Clip and Microsoft SenseCam), are commonly worn on the chest, and are characterized by a very low frame rate (up to 2fpm) that allows to capture images over a long period of time without the need of recharging the battery. Consequently, they offer considerable potential for inferring knowledge about e.g. behaviour patterns, habits or lifestyle of the user. However, due to the low frame-rate and the free motion of the camera, temporally adjacent images typically present abrupt appearance changes so that motion features cannot be reliably estimated.
• The latter (e.g., Google Glass, GoPro), are commonly mounted on the head, and capture conventional video (around 35fps) that allows to capture fine temporal details of interactions. Consequently, they offer potential for in-depth analysis of daily or special activities. However, since the camera is moving with the wearer head, it becomes more difficult to estimate the global motion of the wearer and in the case of abrupt movements, the images can result blurred.

In both cases, since the camera is worn in a naturalistic setting, visual data present a huge variability in terms of illumination conditions and object appearance.Moreover, the camera wearer is not visible in the image and what he/she is doing has to be inferred from the information in the visual field of the camera, implying that important information about the wearer, such for instance as pose or facial expression estimation, is not available.

Applications

A collection of studies published in a special theme issue of the American Journal of Preventive Medicine^[6] has demonstrated the potential of lifelogs captured through wearable cameras from a number of viewpoints. In particular, it has been shown that used as a tool for understanding and tracking lifestyle behaviour, lifelogs would enable the prevention of noncommunicable diseases associated to unhealthy trends and risky profiles (such as obesity, depression, etc.). In addition, used as a tool of re-memory cognitive training, lifelogs would enable the prevention of cognitive and functional decline in elderly people.

More recently, egocentric cameras have been used to study human and animal cognition, human-human social interaction, human-robot interaction, human expertise in complex tasks.Other applications include navigation/assistive technologies for the blind,^[20] monitoring and assistance of industrial workflows,^{[21] [22]} and augmented reality interfaces.^[5]

See also

• Eye tracking
• Lifelog
• Quantified self
• Smartglasses
• Sousveillance

Notes and References

1. An Introduction to the 3rd Workshop on Egocentric (First-person) Vision, Steve Mann, Kris M. Kitani, Yong Jae Lee, M. S. Ryoo, and Alireza Fathi, IEEE Conference on Computer Vision and Pattern Recognition Workshops 2160-7508/14, 2014, IEEE
2. Mann, S. (1998). Humanistic computing:" WearComp" as a new framework and application for intelligent signal processing. Proceedings of the IEEE, 86(11), 2123-2151.
3. Haykin, Simon S., and Bart Kosko. Intelligent signal processing. Wiley-IEEE Press, 2001.
4. “Wearable AI”, Steve Mann, Li-Te Cheng, John Robinson, Kaoru Sumi, Toyoaki Nishida, Soichiro Matsushita, Ömer Faruk Özer, Oguz Özun, C. Öncel Tüzel, Volkan Atalay, A. Enis Cetin, Joshua Anhalt, Asim Smailagic, Daniel P. Siewiorek, Francine Gemperle, Daniel Salber, Weber, Jim Beck, Jim Jennings, and David A. Ross, IEEE Intelligent Systems 16(3), 2001, Pages 0(cover) to 53.
5. Book: Mann, S.. Digest of Papers. Fourth International Symposium on Wearable Computers . Telepointer: Hands-free completely self-contained wearable visual augmented reality without headwear and without any infrastructural reliance . October 2000. https://ieeexplore.ieee.org/document/888489. 177–178. 10.1109/ISWC.2000.888489. 0-7695-0795-6 . 6036868 .
6. Doherty, A. R., Hodges, S. E., King, A. C., Smeaton, A. F., Berry, E., Moulin, C. J., ... & Foster, C. (2013). Wearable cameras in health. American Journal of Preventive Medicine, 44(3), 320-323.
7. Bolanos, M., Dimiccoli, M., & Radeva, P. (2017). Toward storytelling from visual lifelogging: An overview. IEEE Transactions on Human-Machine Systems, 47(1), 77-90.
8. Kanade. Takeo. Hebert. Martial. August 2012. First-Person Vision. Proceedings of the IEEE. 100. 8. 2442–2453. 10.1109/JPROC.2012.2200554. 33060600 . 1558-2256.
9. Fathi, A., Farhadi, A., & Rehg, J. M. (2011, November). Understanding egocentric activities. In Computer Vision (ICCV), 2011 IEEE International Conference on (pp. 407-414). IEEE.
10. Fathi, A., Hodgins, J. K., & Rehg, J. M. (2012, June). Social interactions: A first-person perspective. In Computer Vision and Pattern Recognition (CVPR), 2012 IEEE Conference on (pp. 1226-1233). IEEE.
11. Poleg, Y., Arora, C., & Peleg, S. (2014). Temporal segmentation of egocentric videos. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (pp. 2537-2544).
12. Lee, Y. J., Ghosh, J., & Grauman, K. (2012, June). Discovering important people and objects for egocentric video summarization. In Computer Vision and Pattern Recognition (CVPR), 2012 IEEE Conference on (pp. 1346-1353). IEEE.
13. Park, H. S., Jain, E., & Sheikh, Y. (2012). 3d social saliency from head-mounted cameras. In Advances in Neural Information Processing Systems (pp. 422-430).
14. Book: Su. Yu-Chuan. Grauman. Kristen. Computer Vision – ECCV 2016 . Detecting Engagement in Egocentric Video . 2016. Leibe. Bastian. Matas. Jiri. Sebe. Nicu. Welling. Max. https://link.springer.com/chapter/10.1007/978-3-319-46454-1_28. Lecture Notes in Computer Science . 9909 . en. Cham. Springer International Publishing. 454–471. 10.1007/978-3-319-46454-1_28. 978-3-319-46454-1. 1604.00906. 1599840 .
15. Rogez, G., Supancic, J. S., & Ramanan, D. (2015). First-person pose recognition using egocentric workspaces. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 4325-4333).
16. Mann, S., Janzen, R., Ai, T., Yasrebi, S. N., Kawwa, J., & Ali, M. A. (2014, May). Toposculpting: Computational lightpainting and wearable computational photography for abakographic user interfaces. In Electrical and Computer Engineering (CCECE), 2014 IEEE 27th Canadian Conference on (pp. 1-10). IEEE.
17. Bettadapura, V., Essa, I., & Pantofaru, C. (2015, January). Egocentric field-of-view localization using first-person point-of-view devices. In Applications of Computer Vision (WACV), 2015 IEEE Winter Conference on (pp. 626-633). IEEE
18. Ji. Peng. Song. Aiguo. Xiong. Pengwen. Yi. Ping. Xu. Xiaonong. Li. Huijun. 2017-09-01. Egocentric-Vision based Hand Posture Control System for Reconnaissance Robots. Journal of Intelligent & Robotic Systems. en. 87. 3. 583–599. 10.1007/s10846-016-0440-2. 254648250 . 1573-0409.
19. Book: Bokhari. Syed Zahir. Kitani. Kris M.. Computer Vision – ACCV 2016 . Long-Term Activity Forecasting Using First-Person Vision . 2017. Lai. Shang-Hong. Lepetit. Vincent. Nishino. Ko. Sato. Yoichi. https://link.springer.com/chapter/10.1007/978-3-319-54193-8_22. Lecture Notes in Computer Science . 10115 . en. Cham. Springer International Publishing. 346–360. 10.1007/978-3-319-54193-8_22. 978-3-319-54193-8.
20. Yagi, T., Mangalam, K., Yonetani, R., & Sato, Y. (2017). Future Person Localization in First-Person Videos. arXiv preprint .
21. Book: Leelasawassuk. Teesid. Damen. Dima. Mayol-Cuevas. Walterio. Proceedings of the 8th Augmented Human International Conference . Automated capture and delivery of assistive task guidance with an eyewear computer . 2017-03-16. https://doi.org/10.1145/3041164.3041185. AH '17. New York, NY, USA. Association for Computing Machinery. 1–9. 10.1145/3041164.3041185. 1983/ed89a4ab-f375-40b7-bdf4-b3f97925a0fe . 978-1-4503-4835-5. 10231349 .
22. Edmunds, S. R., Rozga, A., Li, Y., Karp, E. A., Ibanez, L. V., Rehg, J. M., & Stone, W. L. (2017). Brief Report: Using a Point-of-View Camera to Measure Eye Gaze in Young Children with Autism Spectrum Disorder During Naturalistic Social Interactions: A Pilot Study. Journal of Autism and Developmental Disorders, 47(3), 898-904.