First-Person Activity Recognition: What Are They Doing to Me?

By Michael S. Ryoo, Larry H. Matthies. 2013

316 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;…”
Ego4D: Around the World in 3,000 Hours of Egocentric Video (2021, background): “Furthermore, while existing datasets rely largely on graduate students as camera wearers [43,44,66,129,129,138, 168, 179, 194, 210], Ego4D camera wearers are of a much wider demographic, as detailed below.”
Delving into Egocentric Actions (2015, background): “Ryoo and Matthies [31] combine a global motion descriptor with local motion cues for interaction-level action recognition.”
Scaling Egocentric Vision: The EPIC-KITCHENS Dataset (2018, background): “We exclude egocentric datasets that focus on inter-person interactions [27], [28], [29] as well as instructional videos [30], [31], as these target a different research question.”
Charades-Ego: A Large-Scale Dataset of Paired Third and First Person Videos (2018, background): “In recent years, the field of egocentric action understanding [4, 5, 6, 8, 3, 7] has exploded due to a variety of applications, including augmented and virtual reality.”
Multi-Stream Deep Neural Networks for RGB-D Egocentric Action Recognition (2019, unknown): “JPL-Interaction [41] 1 62 RGB Head – 7 2013 ! #”
Predicting Important Objects for Egocentric Video Summarization (2015, unknown): “There are four main steps to our approach: (1) us-ing novel egocentric saliency cues to train a category-independent regression model that predicts how likely it is that an image region belongs to an important person or object; (2) partitioning the video into temporal events.”
First-Person Activity Forecasting with Online Inverse Reinforcement Learning (2016, background): “First-person vision (FPV): Wearable cameras have been used for various human behavior understanding tasks [4, 10, 12, 16, 22] because they give direct access to detailed visual information about a person’s actions.”
Pooled motion features for first-person videos (2014, background): “Most of the previous firstperson activity recognition works [9, 18] focused on the use of existing features and representations designed for conventional 3rd-person videos, without tailoring motion features for the first-person case.”
Next-active-object prediction from egocentric videos (2017, background): “Ryoo and Matthies [30] considered videos acquired from a robot-centric perspective and proposed to recognize egocentric activities performed by other subjects while interacting with the robot.”
Egocentric Future Localization (2016, background): “This characteristics provides a powerful cue to understand human behaviors [6, 8, 15, 34, 36].”
Part-Activated Deep Reinforcement Learning for Action Prediction (2018, background): “The majority of the existing action prediction works can be mainly divided into two categories: exploiting reliable features with template matching [2,20,44] and developing classification models [14,16,31,33].”
Visual Motif Discovery via First-Person Vision (2016, background): “Because many prior studies on first-person vision focused on how their method could be generalized to a variety of environments (e.g., GeorgiaTech Egocentric Activities [4], JPL First-Person Interaction [6], CMU Social Saliency [29]), there are few datasets of first-person videos that have been reco”
Multitype Activity Recognition in Robot-Centric Scenarios (2015, background): “In the last few years, also first-person activities [5], [18], [4] and firstperson interactions [6], [7] have been analyzed.”
This Hand Is My Hand: A Probabilistic Approach to Hand Disambiguation in Egocentric Video (2014, background): “Egocentric video is becoming a popular research topic in computer vision, with papers recognizing activities [7, 8, 19, 22], events [15, 18, 25], and objects [9, 21].”
Recognizing Micro-Actions and Reactions from Paired Egocentric Videos (2016, methodology): “Particularly, Ryoo and Matthies have addressed the problem of recognizing interactions from egocentric videos [29].”
Jointly Learning Energy Expenditures and Activities Using Egocentric Multimodal Signals (2017, background): “[23, 20] reason about state changes in household objects, and [17, 54] reason about human-object interactions.”
The THUMOS challenge on action recognition for videos "in the wild" (2016, unknown): “Recent datasets contain challenging videos with actions from various sources such as movies ( Kuehne et al., 2011; Marszałek et al., 2009 ), YouTube ( Liu et al., 2009 ), and wearable cameras ( Pirsiavash and Ramanan, 2012; Ryoo and Matthies, 2013 ).”