Attention-Aware Deep Reinforcement Learning for Video Face Recognition

By Yongming Rao, Jiwen Lu, Jie Zhou. 2017

171 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Reinforcement Learning

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

1. Xiaogang Wang
2. Dacheng Tao
3. Luc Van Gool
4. Xiaoou Tang
5. Chen Change Loy
6. Jie Zhou
7. Jiwen Lu
8. Ming-Hsuan Yang

In Reinforcement Learning

1. Gao Huang
2. Jiwen Lu
3. Jie Zhou
4. Xiangyang Ji
5. Jian Cheng
6. Hanqing Lu
7. Yifan Zhang

Citations in the corpus

• Dynamic Neural Networks: A Survey (2021, methodology): “To improve the inference efficiency, hard attention is realized to remove unimportant frames iteratively with RL for efficient video face verification [188].”
• Discriminability Distillation in Group Representation Learning (2020, methodology): “p-based recognition is how to dene an ecient indicator to select representatives from sets. To tackle such cases, previous methods aim at dening the quality or saliency for each element in a group [37,62,42,40]. The weight for each element can be automatically learned by self-attention. For example,”
• TransMIL: Transformer based Correlated Multiple Instance Learning for Whole Slide Image Classication (2021, methodology): “Then the attention mechanism was gradually applied to computer vision tasks, including giving different weights to feature channels [30] or spatial distribution [31], or giving different weights to time series in video analysis [32].”
• Counterfactual Attention Learning for Fine-Grained Visual Categorization and Re-identification (2021, unknown): “visual categories, visual attention mechanism has proven to be especially effective in fine-grained visual recognition tasks and become a core component in many state-of-the-art methods [47, 31, 7, 65, 50, 19, 28, 25, 62, 44].”
• Neural Aggregation Network for Video Face Recognition (2016, background): “So far, a variety of efforts on integrating information across different frames have been dedicated [18, 25, 27, 42, 28, 11, 32, 7, 8, 1].”
• Deep Progressive Reinforcement Learning for Skeleton-Based Action Recognition (2018, background): “More specifically, [37] decided whether to drop a frame or not and [34] selected one frame at each step, while we deal with the adjustments of all the selected frames at one time.”