Research map / Video / Action recognition
Tube Convolutional Neural Network (T-CNN) for Action Detection in Videos
Rui Hou, Chen Chen, Mubarak Shah
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 temporal skip pooling (ablation) | T-CNN · J-HMDB | frame-mAP (IoU 0.5): 47.9 -> 61.3 | Improved |
| add temporal skip pooling (ablation) | T-CNN · J-HMDB | video-mAP (IoU 0.2): 66.9 -> 78.4 | Improved |
| add temporal skip pooling (ablation) | T-CNN · J-HMDB | video-mAP (IoU 0.5): 58.6 -> 76.9 | Improved |
| concatenate conv2 feature with conv5 (skip pooling) (ablation) | T-CNN TPN · UCF-Sports | frame-mAP (IoU 0.5): 74.9 -> 86.7 | Improved |
| concatenate conv2 feature with conv5 (skip pooling) (ablation) | T-CNN TPN · UCF-Sports | video-mAP (IoU 0.5): 77.9 -> 94.8 | Improved |
| skip from conv1 instead of conv2 (ablation) | T-CNN TPN · UCF-Sports | frame-mAP (IoU 0.5): 86.7 -> 81.2 | Hurt |
| skip from conv3 instead of conv2 (ablation) | T-CNN TPN · UCF-Sports | video-mAP (IoU 0.5): 94.8 -> 91.7 | Hurt |
| skip from conv4 instead of conv2 (ablation) | T-CNN TPN · UCF-Sports | frame-mAP (IoU 0.5): 86.7 -> 77.6 | Hurt |
| replace clip-wise C3D pool5 with ToI pooling | C3D · UCF-101 (101 actions) | accuracy (%): 82.3 -> 87.5 | Improved |
| T-CNN versus Peng et al. detection | T-CNN · UCF-Sports | frame-mAP (IoU 0.5): 84.51 -> 86.7 | Improved |
| T-CNN versus Peng et al. detection | T-CNN · J-HMDB | video-mAP (IoU 0.5): 73.1 -> 76.9 | Improved |
| T-CNN versus Peng et al. detection | T-CNN · UCF-101 (24 actions) | video-mAP (IoU 0.2): 42.3 -> 47.1 | Improved |
Closest papers by meaning
- AMTnet: Action-Micro-Tube Regression by End-To-End Trainable Deep Architecture (ICCV 2017 · similarity 0.90)
- Recurrent Tubelet Proposal and Recognition Networks for Action Detection (ECCV 2018 · similarity 0.89)
- T-C3D: Temporal Convolutional 3D Network for Real-Time Action Recognition (AAAI 2018 · similarity 0.88)
- Object Detection From Video Tubelets With Convolutional Neural Networks (CVPR 2016 · similarity 0.88)
- TubeR: Tubelet Transformer for Video Action Detection (CVPR 2022 · similarity 0.88)
- Temporal Convolutional Networks for Action Segmentation and Detection (CVPR 2017 · similarity 0.87)
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.
