Research map / Video / Action recognition
Deep Discriminative Model for Video Classification
Mohammad Tavakolian, Abdenour Hadid
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 SCSP features as input instead of raw video (ablation) | DDM · UCF101 | accuracy (%): 91.5 -> 94.3 | Improved |
| add SCSP features (ablation) | DDM · Hollywood II | mAP (%): 75.3 -> 77.1 | Improved |
| add SCSP features (ablation) | DDM · DynTex | rank-1 recognition rate (%): 98.05 -> 99.27 | Improved |
| add SCSP features (ablation) | DDM · YUPENN | accuracy (%): 97.52 -> 99.18 | Improved |
| add SCSP features (ablation) | DDM · Maryland | accuracy (%): 86.33 -> 90.27 | Improved |
| use SCSP block size 3x3x3 instead of 1x1x3 (ablation) | DDM+SCSP · UCF101 | accuracy (%): 87.3 -> 94.3 | Improved |
| use SCSP block size 3x3x3 instead of 7x7x5 (ablation) | DDM+SCSP · UCF101 | accuracy (%): 79.5 -> 94.3 | Improved |
| use SCSP block size 3x3x3 instead of 3x3x5 (ablation) | DDM+SCSP · UCF101 | accuracy (%): 92.5 -> 94.3 | Improved |
| use SCSP block size 3x3x3 instead of 5x5x5 (ablation) | DDM+SCSP · UCF101 | accuracy (%): 84.3 -> 94.3 | Improved |
| compare DDM raw with KVMF baseline | DDM · UCF101 | accuracy (%): 93.1 vs 91.5 | Hurt |
| compare DDM+SCSP with st-TCoF on Maryland | DDM+SCSP · Maryland | accuracy (%): 88.46 -> 90.27 | Improved |
| compare DDM+SCSP with HRP baseline | DDM+SCSP · Hollywood II | mAP (%): 76.7 -> 77.1 | Improved |
Closest papers by meaning
- Sympathy for the Details: Dense Trajectories and Hybrid Classification Architectures for Action Recognition (ECCV 2016 · similarity 0.86)
- Discriminative Video Representation Learning Using Support Vector Classifiers (TPAMI 2019 · similarity 0.86)
- Deep Analysis of CNN-Based Spatio-Temporal Representations for Action Recognition (CVPR 2021 · similarity 0.85)
- Video Representation Learning Using Discriminative Pooling (CVPR 2018 · similarity 0.85)
- Harnessing Object and Scene Semantics for Large-Scale Video Understanding (CVPR 2016 · similarity 0.84)
- Revisiting Classifier: Transferring Vision-Language Models for Video Recognition (AAAI 2023 · similarity 0.84)
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.
