Domain-Adaptive Discriminative One-Shot Learning of Gestures

By Tomas Pfister, James Charles, Andrew Zisserman. 2014

94 citations recorded by the source corpus.

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

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

In Signal Processing

Citations in the corpus

Modeling Temporal Dynamics and Spatial Configurations of Actions Using Two-Stream Recurrent Neural Networks (2017, neutral): “Precision Recall F1-score Skeleton Feature [48] 59.9 59.3 59.6 Portfolios [52] – – 56.0 Gesture Spotting [35] 61.2 62.3 61.7 HiVideoDarwin [45] 74.9 75.6 74.6 CNN for Skeleton [5] 91”
Generalizing from a Few Examples (2019, background): “a sources has been recently presented in [119] as a new benchmark. Further, more computer vision applications can be explored, such as image retrieval [118], object tracking [15], gesture recognition [88], image captioning and visual question answering [28] and video event detection [137]. Apart fro”
Discover and Learn New Objects from Documentaries (2017, background): “Several studies [6, 8, 31] automatically learn British Sign Language signs from TV broadcasts.”
Adversarially Robust Few-Shot Learning: A Meta-Learning Approach (2019, background): “When data is scarce or new classes arise frequently, neural networks must adapt quickly [7, 14, 24, 30].”
Two streams Recurrent Neural Networks for Large-Scale Continuous Gesture Recognition (2016, methodology): “[16] proposed a method to boost the performance of one-shot learning by using weakly supervised learning.”
A Deep Neural Framework for Continuous Sign Language Recognition by Iterative Training (2019, background): “These features characterize handshape, appearance and motion cues, by using image pixel intensity [16], gradients [11], [15], [23] and motion trajectories or velocities [8], [11], [17].”
Robust Few-Shot Learning with Adversarially Queried Meta-Learners (2019, background): “When data is scarce or new classes arise frequently, neural networks must adapt quickly (Duan et al., 2017; Kaiser et al., 2017; Pfister et al., 2014; Vartak et al., 2017).”
A Comprehensive Survey of Few-shot Learning: Evolution, Applications, Challenges, and Opportunities (2022, background): “DAD [62] ECCV’14 Data Expansion Sign Language* ✓ ✓ FTT [60] CVPR’16 Data Expansion TADB ✓”
Explore Efficient Local Features from RGB-D Data for One-Shot Learning Gesture Recognition (2016, unknown): “However, the traditional methods are based on strongly supervised learning [8] and require a large number of training samples [9], [10].”
Modeling video evolution for action recognition (2015, unknown): “[23] 61.”
Skeleton based action recognition with convolutional neural network (2015, methodology): “We follow the experimental protocol adopted in [3, 14, 22, 25] and provide precision, recall and F1-score measures on the validation set.”
SemGIF: A Semantics Guided Incremental Few-shot Learning Framework with Generative Replay (2021, background): “The impact of alleviating the data gathering effort can be widely seen from the plethora of fields benefited from the same like image recognition [39], image retrieval [38] and gesture recognition [24].”
Hierarchical motion evolution for action recognition (2015, methodology): “We report precision, recall, F1-score and mAP on the validation set, as done in [17, 28].”