Research map / Detection and segmentation / CNN features
DeepCAMP: Deep Convolutional Action & Attribute Mid-Level Patterns
Ali Diba, Ali Mohammad Pazandeh, Hamed Pirsiavash, Luc Van Gool
Check 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 iterative patch-cluster learning and update the representation and clusters (ablation) | DeepCAMP with AlexNet patch CNN · PASCAL VOC 2012 action validation | mAP: 68.4 -> 71.7 | Improved |
| Continue PatternNet iterative learning from 1 to 2 iterations (ablation) | DeepCAMP · PASCAL VOC 2012 action validation | mAP: 73.5 -> 75.3 | Improved |
| Continue PatternNet iterative learning from 2 to 3 iterations (ablation) | DeepCAMP · PASCAL VOC 2012 action validation | mAP: 75.3 -> 75.4 | No clear effect |
| Use PatternNet instead of AlexNet as the patch CNN (ablation) | DeepCAMP · PASCAL VOC 2012 action validation | mAP: 71.7 -> 75.4 | Improved |
| Use PatternNet instead of AlexNet as the patch CNN (ablation) | DeepCAMP · PASCAL VOC 2012 action test | mAP: 73.9 -> 77.6 | Improved |
| Use PatternNet instead of AlexNet as the patch CNN (ablation) | DeepCAMP · Stanford40 | AP: 49 -> 52.6 | Improved |
| Use PatternNet instead of AlexNet as the patch CNN (ablation) | DeepCAMP · Berkeley Attributes of People | mAP: 84.1 -> 86.6 | Improved |
| Add MDPM features to the CNN representation | CNN action baseline · PASCAL VOC 2012 action validation | mAP: 68.4 -> 69 | Improved |
| Add MDPM features to the attribute CNN | CNN_att · Berkeley Attributes of People | mAP: 80.6 -> 81.9 | Improved |
| Use the full PatternNet pipeline rather than the strongest listed previous method | DeepCAMP · PASCAL VOC 2012 action test | mAP: 76.3 -> 77.6 | Improved |
| Use the full PatternNet pipeline rather than the strongest listed previous method | DeepCAMP · Berkeley Attributes of People | mAP: 86 -> 86.6 | Improved |
Closest papers by meaning
- Sympathy for the Details: Dense Trajectories and Hybrid Classification Architectures for Action Recognition (ECCV 2016 · similarity 0.87)
- Attributes for Improved Attributes: A Multi-Task Network Utilizing Implicit and Explicit Relationships for Facial Attribute Classification (AAAI 2017 · similarity 0.86)
- Discriminative Video Representation Learning Using Support Vector Classifiers (TPAMI 2019 · similarity 0.86)
- Temporal Convolutional Networks for Action Segmentation and Detection (CVPR 2017 · similarity 0.86)
- Deep Alternative Neural Network: Exploring Contexts as Early as Possible for Action Recognition (NeurIPS 2016 · similarity 0.86)
- Deep Analysis of CNN-Based Spatio-Temporal Representations for Action Recognition (CVPR 2021 · similarity 0.86)
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.
