Adversarial Cross-Domain Action Recognition with Co-Attention

By Boxiao Pan, Zhangjie Cao, Ehsan Adeli, Juan Carlos Niebles. 2019

116 citations recorded by the source corpus.

Action recognition has been a widely studied topic with a heavy focus on supervised learning involving sufficient labeled videos. However, the problem of cross-domain action recognition, where training and testing videos are drawn from different underlying distributions, remains largely under-explored. Previous methods directly employ techniques for cross-domain image recognition, which tend to suffer from the severe temporal misalignment problem. This paper proposes a Temporal Co-attention Network (TCoN), which matches the distributions of temporally aligned action features between source and target domains using a novel cross-domain co-attention mechanism. Experimental results on three cross-domain action recognition datasets demonstrate that TCoN improves both previous single-domain and cross-domain methods significantly under the cross-domain setting.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Safety, Robustness & Interpretability

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

1. Luc Van Gool
2. Giovanni Maria Farinella
3. Antonino Furnari
4. Wenguan Wang
5. Guolei Sun
6. Dima Damen
7. Davide Moltisanti
8. Jifeng Dai

In Computer Vision and Pattern Recognition

1. Kristen Grauman
2. Giovanni Maria Farinella
3. Antonino Furnari
4. Luc Van Gool
5. Jitendra Malik
6. Cordelia Schmid
7. Andrew Zisserman
8. Dima Damen

In Safety, Robustness & Interpretability

1. Dima Damen
2. Luc Van Gool
3. Wenguan Wang
4. Jifeng Dai
5. Jonathan Munro

Citations in the corpus

• Rescaling Egocentric Vision: Collection, Pipeline and Challenges for EPIC-KITCHENS-100 (2020, background): “frames [112], and avoiding non-informative background frames [114].”
• Multi-Modal Domain Adaptation for Fine-Grained Action Recognition (2020, methodology): “Three recent works attempted deep UDA [7, 19, 36].”
• Mining Cross-Image Semantics for Weakly Supervised Semantic Segmentation (2020, background): “When Im and In are from different domains, the parameter matrix WP , in essence, learns to map them into a unified common semantic space [44] and the co-attentive features can capture domainshared semantics.”
• Multi-Modal Domain Adaptation for Fine-Grained Action Recognition (2019, methodology): “Three recent works attempted deep UDA [7, 19, 36].”