Research map / Video / Action recognition
TORNADO: A Spatio-Temporal Convolutional Regression Network for Video Action Proposal
Hongyuan Zhu, Romain Vial, Shijian Lu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add T-CRN (ConvLSTM) to S-CRN on RGB stream (ablation) | TORNADO RGB · UCF-101 | recall at IoU 0.5: 56.10 -> 59.59 | Improved |
| add T-CRN to S-CRN on RGB stream (ablation) | TORNADO RGB · UCF-Sports | recall at IoU 0.5: 96.79 -> 96.79 | No clear effect |
| add S-CRN to T-CRN on Flow stream (ablation) | TORNADO Flow · UCF-101 | recall at IoU 0.5: 47.29 -> 49.70 | Improved |
| add S-CRN to T-CRN on Flow stream (ablation) | TORNADO Flow · UCF-Sports | recall at IoU 0.5: 92.54 -> 94.67 | Improved |
| add S-CRN to T-CRN on RGB stream (ablation) | TORNADO RGB · UCF-101 | recall at IoU 0.5: 55.10 -> 59.59 | Improved |
| add Flow stream to RGB stream (ablation) | TORNADO · UCF-101 | recall at IoU 0.5: 59.59 -> 61.42 | Improved |
| add Flow stream to RGB stream (ablation) | TORNADO · UCF-Sports | recall at IoU 0.5: 96.79 -> 96.79 | No clear effect |
| add Flow stream to RGB stream (ablation) | TORNADO · UCF-Sports | MABO: 74.26 -> 76.20 | Improved |
| add peak-pattern path trimming (ablation) | TORNADO-RGB+OF · UCF-101 | recall at IoU 0.5: 57.92 -> 61.42 | Improved |
| add peak-pattern path trimming (ablation) | TORNADO-RGB+OF · UCF-101 | MABO: 50.71 -> 54.93 | Improved |
| add peak-pattern path trimming (ablation) | TORNADO-RGB+OF · UCF-Sports | MABO: 76.20 -> 76.20 | No clear effect |
| add T-CRN to S-CRN on RGB stream (ablation) | TORNADO RGB · UCF-Sports | MABO: 73.83 -> 74.26 | Improved |
Closest papers by meaning
- Temporal Convolutional Networks for Action Segmentation and Detection (CVPR 2017 · similarity 0.90)
- Learning Motion in Feature Space: Locally-Consistent Deformable Convolution Networks for Fine-Grained Action Detection (ICCV 2019 · similarity 0.89)
- Spatiotemporal Residual Networks for Video Action Recognition (NeurIPS 2016 · similarity 0.89)
- Convolutional Two-Stream Network Fusion for Video Action Recognition (CVPR 2016 · similarity 0.88)
- StNet: Local and Global Spatial-Temporal Modeling for Action Recognition (AAAI 2019 · similarity 0.88)
- Recurrent Tubelet Proposal and Recognition Networks for Action Detection (ECCV 2018 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
