Task-Specific Gesture Analysis in Real-Time Using Interpolated Views

By Trevor Darrell, Irfan Essa, Alex Pentland. 1996

How researchers see this paper

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

Citations in the corpus

Histogram of Oriented Principal Components for Cross-View Action Recognition (2014, background): “Since HOPC and STK-Distribution capture complimentary information, their combination significantly improves classification accuracy which is over 22% higher than the nearest competitor on both datasets.”
HOPC: Histogram of Oriented Principal Components of 3D Pointclouds for Action Recognition (2014, methodology): “Some methods infer the 3D scene structure and use geometric transformations to achieve view invariance [5,9,15,26,39].”
Challenges in multimodal gesture recognition (2016, methodology): “Most indirect methods (Alon et al., 2009; Lee and Kim, 1999; Oka, 1998) are based on extensions of Dynamic Programming (DP) e.g., Dynamic Time Warping (DTW) (Darrell et al., 1996; Kruskal and Liberman, 1983), Continuous Dynamic Programming (CDP) (Oka, 1998), various forms of Hidden Markov Models (HM”
Learning a non-linear knowledge transfer model for cross-view action recognition (2015, background): “These methods often require robust joint estimation which is still an open problem in real-world settings.”
Learning Individual Styles of Conversational Gesture (2019, background): “There has also been work that recognizes emblematic hand and face gestures [17, 14], head gestures [35], and co-speech gestures [38].”
Machine Recognition of Human Activities: A Survey (2008, methodology): “To model highly non-linea r warping functions, the most common method is Dynamic Time Warping (DTW) of the feature sequence such as in [157], [148], [158], [159].”
Learning a Deep Model for Human Action Recognition from Novel Viewpoints (2016, methodology): “To overcome this problem, some techniques infer 3D scene structure and use geometric transformations to achieve view invariance [3], [24], [25], [26], [27], [28], [29], [30], [31], [32].”
Looking at People: Sensing for Ubiquitous and Wearable Computing (2000, methodology): “Once the target classes' PDF has been learned, we can use Bayes' rule to perform maximum a posteriori (MAP) detection and recognition.”
A Unified Framework for Gesture Recognition and Spatiotemporal Gesture Segmentation (2009, methodology): “The digit prototype was the training example for that digit that had the smallest sum of DTW distances to all other training examples for that digit.”