Fast unsupervised ego-action learning for first-person sports videos

By Kris M. Kitani, Takahiro Okabe, Yoichi Sato, Akihiro Sugimoto. 2011

265 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

Citations in the corpus

Socratic Models: Composing Zero-Shot Multimodal Reasoning with Language (2022, background): “This includes ego-motion (e.g., optical flows) (Kitani et al. 2011; Ryoo and Matthies 2013) as well as features from human gaze, hands, and objects (Spriggs, De La Torre, and Hebert 2009; Lee, Ghosh, and Grauman 2012; Fathi, Farhadi, and Rehg 2011; Pirsiavash and Ramanan 2012; Li and Kitani 2013;…”
Delving into Egocentric Actions (2015, background): “[15] track camera motion using sparse optical flow, and encode the motion of tracked points into a histogram, which is used to cluster egocentric video content and discover egocentric actions.”
LSTM-CNN Architecture for Human Activity Recognition (2020, background): “[11] presented a sparse optical flow algorithm to acquire the histogram of human motion features and proposed an unsupervised Dirichley hybrid model to classify 11 human activities.”
Going Deeper into First-Person Activity Recognition (2016, methodology): “[14] used optical flowbased global motion descriptors to discover ego-action in sports videos.”
Next-active-object prediction from egocentric videos (2017, background): “[28] addressed the problem of discovering egocentric action categories from first person sports videos in an unsupervised 85”
Multi-Stream Deep Neural Networks for RGB-D Egocentric Action Recognition (2019, unknown): “UEC EgoAction [39] 1 2 RGB Head – 37 2011 ! #”
Predicting Important Objects for Egocentric Video Summarization (2015, methodology): “We ﬁrst describe how we collect the video data and ground-truth annotations needed to train our model.”
In the Eye of the Beholder: Gaze and Actions in First Person Video (2020, methodology): “[6] used a global motion descriptor to discover egocentric actions.”
Learning to Recognize Daily Actions Using Gaze (2012, background): “First-Person Vision: The idea of using wearable cameras is not new [22], however, recently there has been a growing interest in using them in the computer vision community, motivated by the advances in hardware technology [23,5,13,21,28,7,6,15,19].”
Jointly Learning Energy Expenditures and Activities Using Egocentric Multimodal Signals (2017, methodology): “[33] introduces unsupervised ego-action discovery using Dirichlet processes.”
H+O: Unified Egocentric Recognition of 3D Hand-Object Poses and Interactions (2019, background): “Early approaches for egocentric action recognition rely on motion cues [29, 50, 55].”
Egocentric Future Localization (2016, background): “This characteristics provides a powerful cue to understand human behaviors [6, 8, 15, 34, 36].”
The Open World of Micro-Videos (2016, background): “Wearable cameras such as Google Glass and GoPro have spurred interest in analyzing egocentric views [18,19,20,21].”
First-Person Activity Recognition: What Are They Doing to Me? (2013, background): “It is different from [6, 4, 10] as well, in the aspect that our videos contain movements of interacting persons as well as ego-motion of the observer.”
Recognizing Micro-Actions and Reactions from Paired Egocentric Videos (2016, background): “In this way, egocentric videos are advantageous from a sensing perspective since the head motion and hand motion of camera wearers are often observed clearly in such videos, making it possible to perform various forms of first-person action recognition [7, 9, 13, 18, 25, 27, 30].”
Temporal Perception and Prediction in Ego-Centric Video (2015, background): “Quite a few previous works have used techniques such as, hand detection, object detection, or motion features such as optical flow to represent ego-centric videos [30, 9, 18, 33, 10], but how to represent first person videos effectively for temporal prediction tasks is an open question.”
Summarization of Egocentric Videos: A Comprehensive Survey (2017, background): “[53], which is a compilation of short videos from YouTube and recordings of choreographed activities; or SumMe [17], in which even if the egocentric videos are annotated specifically”