Research map / Label-efficient and robust learning / Domain adaptation and generalization
Partial Video Domain Adaptation With Partial Adversarial Temporal Attentive Network
Yuecong Xu, Jianfei Yang, Haozhi Cao, Zhenghua Chen, Qi Li, Kezhi Mao
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 label attention weights (ablation) | PATAN · U-14->H-7 | top-1 accuracy: 71.43 -> 73.81 | Improved |
| add label attention weights (ablation) | PATAN · H-14->U-7 | top-1 accuracy: 85.34 -> 89.85 | Improved |
| add local temporal weights to class weights (ablation) | PATAN · U-14->H-7 | top-1 accuracy: 70.47 -> 73.81 | Improved |
| add local temporal weights to class weights (ablation) | PATAN · H-14->U-7 | top-1 accuracy: 84.21 -> 89.85 | Improved |
| apply class weights to source classifiers (ablation) | PATAN · U-14->H-7 | top-1 accuracy: 69.52 -> 73.81 | Improved |
| apply class weights to source classifiers (ablation) | PATAN · H-14->U-7 | top-1 accuracy: 82.71 -> 89.85 | Improved |
| apply class weights to domain discriminators (ablation) | PATAN · U-14->H-7 | top-1 accuracy: 67.14 -> 73.81 | Improved |
| apply class weights to domain discriminators (ablation) | PATAN · H-14->U-7 | top-1 accuracy: 81.58 -> 89.85 | Improved |
| use UDA/VUDA methods (DANN, TA3N, MK-MMD, MCD, MDD) instead of source-only TRN | TRN · six PVDA settings | top-1 accuracy | Hurt |
| use PATAN instead of PADA | PADA · six PVDA settings | top-1 accuracy: 11.27% average relative improvement | Improved |
| reduce number of target classes | DANN, PADA, ETN · UCF-HMDB partial | target accuracy | Hurt |
| reduce number of target classes | PATAN · UCF-HMDB partial | target accuracy | No clear effect |
Closest papers by meaning
- Learning to Transfer Examples for Partial Domain Adaptation (CVPR 2019 · similarity 0.89)
- Discriminative Partial Domain Adversarial Network (ECCV 2020 · similarity 0.89)
- Partial Adversarial Domain Adaptation (ECCV 2018 · similarity 0.88)
- Adversarial Partial Domain Adaptation by Cycle Inconsistency (ECCV 2022 · similarity 0.88)
- Temporal Attentive Alignment for Large-Scale Video Domain Adaptation (ICCV 2019 · similarity 0.87)
- Source-Free Video Domain Adaptation by Learning Temporal Consistency for Action Recognition (ECCV 2022 · 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.
