Loading...
Please wait, while we are loading the content...
Similar Documents
Temporal Localization of Fine-Grained Actions in Videos by Domain Transfer from Web Images
| Content Provider | CiteSeerX |
|---|---|
| Author | Sun, Chen Shetty, Sanketh Sukthankar, Rahul Nevatia, Ram |
| Abstract | We address the problem of fine-grained action localization from temporally untrimmed web videos. We assume that only weak video-level annotations are available for training. The goal is to use these weak labels to identify temporal segments corresponding to the actions, and learn models that generalize to unconstrained web videos. We find that web images queried by action names serve as well-localized highlights for many actions, but are noisily labeled. To solve this problem, we propose a simple yet effective method that takes weak video labels and noisy image labels as in-put, and generates localized action frames as output. This is achieved by cross-domain transfer between video frames and web images, using pre-trained deep convolutional neu-ral networks. We then use the localized action frames to train action recognition models with long short-term mem-ory networks. We collect a fine-grained sports action data set FGA-240 of more than 130,000 YouTube videos. It has 240 fine-grained actions under 85 sports activities. Convinc-ing results are shown on the FGA-240 data set, as well as the THUMOS 2014 localization data set with untrimmed training videos. |
| File Format | |
| Access Restriction | Open |
| Subject Keyword | Web Image Fine-grained Action Domain Transfer Temporal Localization Weak Video Label Weak Video-level Annotation Video Frame Untrimmed Web Video Unconstrained Web Video Action Name Fga-240 Data Set Noisy Image Label Youtube Video Localized Action Frame Action Recognition Model Many Action Fine-grained Sport Action Data Temporal Segment Long Short-term Mem-ory Network Well-localized Highlight Weak Label Localization Data Set Action Frame Effective Method Cross-domain Transfer Untrimmed Training Video Learn Model Pre-trained Deep Convolutional Neu-ral Network Fine-grained Action Localization Convinc-ing Result Sport Activity |
| Content Type | Text |
