Pixel-Level Hand Detection in Ego-centric Videos

By Cheng Li, Kris M. Kitani. 2013

275 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Human-Computer Interaction

How researchers see this paper

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

In Computer Vision and Pattern Recognition

In Human-Computer Interaction

Citations in the corpus

Socratic Models: Composing Zero-Shot Multimodal Reasoning with Language (2022, background): “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; Lee and Grauman 2015).”
Fine-Grained Egocentric Hand-Object Segmentation: Dataset, Model, and Applications (2022, background): “Li and Kitani [31,32] first addressed the hand segmentation problem under various illuminations, and proposed to adaptively select a model that works the best under different illumination scenarios during inference time.”
Future Person Localization in First-Person Videos (2017, background): “Recently, some work has focused on activity recognition [7, 22, 23, 28], activity forecasting [6, 9, 26, 31], person identification [11], gaze anticipation [45] and grasp recognition [3, 4, 21, 35].”
3D Hand Shape and Pose From Images in the Wild (2019, unknown): “g.[23, 5]) as they are sensitive to occlusions.”
Egocentric Hand Track and Object-Based Human Action Recognition (2019, methodology): “Initially, hand detection, segmentation and identification techniques [22]– [24] are developed and their results are utilized to model actions or activities.”
This Hand Is My Hand: A Probabilistic Approach to Hand Disambiguation in Egocentric Video (2014, background): “Other work on egocentric hand detection is motivated by the idea that hands are important for understanding complex object manipulations, gestures, and motor skills [16, 17, 24, 32].”
Depth-Based Hand Pose Estimation: Methods, Data, and Challenges (2015, unknown): “Recently introduced consumer depth cameras have spurred a flurry of new advances [4, 14, 15, 17, 25, 33, 36, 38, 42].”
First Person Action-Object Detection with EgoNet (2016, background): “In particular, egocentric spatial representations [16, 21] have demonstrated stronger predictive power than that of a third person for activity recognition [16, 17], gaze movement [19, 22] and future ego-motion [23].”
Who Let the Dogs Out? Modeling Dog Behavior from Visual Data (2018, background): “, [7, 32, 18, 19]) since we study the dog’s behavior from the perspective of the dog.”
Pixel-Level Hand Detection with Shape-Aware Structured Forests (2014, methodology): “5: Sample Images of GTEA dataset: (left column) original images, (second column) results of Li’s Method [15], and (last column) our results.”
Ego2Hands: A Dataset for Egocentric Two-hand Segmentation and Detection (2020, background): “Pioneering work [13] contributed three egocentric videos (EDSH1, EDSH2 and EDSHkitchen) with varying illumination for binary hand segmentation.”
You2Me: Inferring Body Pose in Egocentric Video via First and Second Person Interactions (2019, background): “Egocentric video analysis Recent egocentric vision work focuses primarily on recognizing objects [13], activities [38, 15, 33, 48, 40, 32, 44, 43, 58], visible hand and arm poses [7, 28, 29, 42], eye gaze [30], or anticipating future camera trajectories [35, 9].”
Analysis of Hand Segmentation in the Wild (2018, background): “Li and Kitani [16] proposed an illumination-aware approach that chooses the best local color feature for each environment using scene-level feature probes.”
Multi-Stream Deep Neural Networks for RGB-D Egocentric Action Recognition (2019, methodology): “labeled images, including 652 images from our THU-READ and 743 images from the EDSH dataset [96].”
Understanding Hand-Object Manipulation with Grasp Types and Object Attributes (2016, methodology): “Following [22], we train a multi-model hand detector composed by a”
Going Deeper into First-Person Activity Recognition (2016, methodology): “(1) Hand segmentation network: For training data, we can either use annotated ground-truth hand masks or output of pixel-level hand detectors like [18].”
Lending A Hand: Detecting Hands and Recognizing Activities in Complex Egocentric Interactions (2015, background): “Our technique achieves significantly better accuracy than the baseline of [21] (0.”