【视频动作识别】Cvpr_2018_Attend and Interact_Higher-Order Object Interactions for Video Understanding
在动作识别领域，人跟物体的交互信息是非常重要的，一个动作的完成包含了多个物体与人的复杂交互。以往的工作很少考虑到这样复杂的交互关系。这篇文章要做的就是对视频中的多个物体的交互建模，利用这样的信息来提高动作识别的准确率。 如何对物体之间的交互建模 物体之间的交互可以通过物体之间的重叠来判断，如果两个物体重叠了，那么说明它们在这个动作发生的时候有一些密切的交互信息。而物体重叠的判断可以通过各自的特征的
相关文章
- 1. 【视频行为识别】ECO：Efficient Convolutional Network for Online Video Understanding
- 2. 视频动作识别--Convolutional Two-Stream Network Fusion for Video Action Recognition
- 3. Video Understanding(视频理解)
- 4. What Makes a Video a Video :Analyzing Temporal Information in Video Understanding Models and Dataset
- 5. Detecting and Recognizing Human-Object Interactions
- 6. 视频动作识别--Two-Stream Convolutional Networks for Action Recognition in Videos
- 7. 视频动作识别--Temporal Segment Networks: Towards Good Practices for Deep Action Recognition
- 8. 图像理解--Detecting and Recognizing Human-Object Interactions
- 9. [IDW-CNN]Learning Object Interactions and Descriptions for Semantic Image Segmentation
- 10. Mining Inter-Video Proposal Relations for Video Object Detection
- 更多相关文章...
- • W3C RDF and OWL 活动 - W3C 教程
- • ionic 加载动作 - ionic 教程
- • RxJava操作符（七）Conditional and Boolean
- • Git可视化极简易教程 — Git GUI使用方法
相关标签/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。
欢迎关注本站公众号,获取更多信息
相关文章
- 1. 【视频行为识别】ECO：Efficient Convolutional Network for Online Video Understanding
- 2. 视频动作识别--Convolutional Two-Stream Network Fusion for Video Action Recognition
- 3. Video Understanding(视频理解)
- 4. What Makes a Video a Video :Analyzing Temporal Information in Video Understanding Models and Dataset
- 5. Detecting and Recognizing Human-Object Interactions
- 6. 视频动作识别--Two-Stream Convolutional Networks for Action Recognition in Videos
- 7. 视频动作识别--Temporal Segment Networks: Towards Good Practices for Deep Action Recognition
- 8. 图像理解--Detecting and Recognizing Human-Object Interactions
- 9. [IDW-CNN]Learning Object Interactions and Descriptions for Semantic Image Segmentation
- 10. Mining Inter-Video Proposal Relations for Video Object Detection