Knowledge Integration Networks for Action Recognition AAAI 2020
AAAI 2020 碼隆科技實驗室+南京大學 1 摘要 在這項工作中，我們提出了用於視頻動作識別的知識集成網絡（KINet）。KINet能夠聚合有意義的上下文特徵，這些特徵對於識別動作非常重要，例如人類信息和場景上下文。我們設計了一個由一個動作識別主分支和兩個輔助分支組成的三分支體系結構，該結構允許模型對動作識別中的人和場景知識進行編碼。我們探索了兩種預先訓練的教師網絡模型來提取人類和場景的知識，
相關文章
- 1. PBRNet:Progressive Boundary Refinement Network for Temporal Action Detection （AAAI 2020）
- 2. Spatiotemporal Residual Networks for Video Action Recognition
- 3. Spatiotemporal Multiplier Networks for Video Action Recognition
- 4. 論文：3D Convolutional Neural Networks for Human Action Recognition
- 5. Action Recognition之SlowFast Networks
- 6. 譯：Two-Stream Convolutional Networks for Action Recognition in Videos.md
- 7. 3D Convolutional Neural Networks for Human Action Recognition
- 8. 譯：Two-stream convolutional networks for action recognition in videos
- 9. Spatial Temporal Graph Convolutional Networks for Skeleton-Based Action Recognition
- 10. 譯：3D Convolutional Neural Networks for Human Action Recognition
- 更多相關文章...
- • Swift for 循環 - Swift 教程
- • Scala for循環 - Scala教程
- • IntelliJ IDEA代碼格式化設置
- • 使用阿里雲OSS+CDN部署前端頁面與加速靜態資源
相關標籤/搜索
每日一句
-
每一个你不满意的现在，都有一个你没有努力的曾经。
歡迎關注本站公眾號,獲取更多信息
相關文章
- 1. PBRNet:Progressive Boundary Refinement Network for Temporal Action Detection （AAAI 2020）
- 2. Spatiotemporal Residual Networks for Video Action Recognition
- 3. Spatiotemporal Multiplier Networks for Video Action Recognition
- 4. 論文：3D Convolutional Neural Networks for Human Action Recognition
- 5. Action Recognition之SlowFast Networks
- 6. 譯：Two-Stream Convolutional Networks for Action Recognition in Videos.md
- 7. 3D Convolutional Neural Networks for Human Action Recognition
- 8. 譯：Two-stream convolutional networks for action recognition in videos
- 9. Spatial Temporal Graph Convolutional Networks for Skeleton-Based Action Recognition
- 10. 譯：3D Convolutional Neural Networks for Human Action Recognition