Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning
Multi-Task Zero-Shot Action Recognition with Prioritised Data Augmentation
Xun Xu, Timothy M. Hospedales, Shaogang Gong
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 |
|---|---|---|---|
| replace single-task ridge regression with multi-task embedding (MTE, latent matching) | zero-shot action recognition · HMDB51 | accuracy: 18.3 -> 19.7 | Improved |
| replace single-task ridge regression with multi-task embedding (MTE, latent matching) | zero-shot action recognition · UCF101 | accuracy: 14.5 -> 15.8 | Improved |
| replace single-task ridge regression with multi-task embedding (MTE, latent matching) | zero-shot action recognition · Olympic Sports | average precision: 40.9 -> 44.3 | Improved |
| add latent-space matching (ablation) | MTE · UCF101 | accuracy: 14.2 -> 15.8 | Improved |
| add latent-space matching (ablation) | MTE · HMDB51 | accuracy: 18.7 -> 19.7 | Improved |
| add latent-space matching (ablation) | MTE · Olympic Sports | average precision: 44.5 -> 44.3 | No clear effect |
| add naive data augmentation from other datasets | RR · HMDB51 (zero-shot, 5 splits) | accuracy: 18.3 -> 21.9 | Improved |
| add naive data augmentation from other datasets | MTE · UCF101 (zero-shot, 5 splits) | accuracy: 15.8 -> 20.9 | Improved |
| add category KLIEP weighting instead of naive DA (ablation) | RR · Olympic Sports (zero-shot, 5 splits) | average precision: 46.5 -> 51.8 | Improved |
| add full KLIEP weighting instead of naive DA (ablation) | MTE · UCF101 (zero-shot, 5 splits) | accuracy: 20.9 -> 21.9 | Improved |
| add full KLIEP weighting instead of naive DA (ablation) | MTE · Olympic Sports (zero-shot, 5 splits) | average precision: 49.4 -> 52.3 | Improved |
| add full KLIEP on top of visual or category weighting alone (ablation) | MTE · HMDB51 (zero-shot, 5 splits) | accuracy: 23.4 -> 23.9 | No clear effect |
Closest papers by meaning
- Transductive Zero-Shot Action Recognition by Word-Vector Embedding (IJCV 2017 · similarity 0.94)
- Cross-Modal Representation Learning for Zero-Shot Action Recognition (CVPR 2022 · similarity 0.92)
- Rethinking Zero-Shot Learning: A Conditional Visual Classification Perspective (ICCV 2019 · similarity 0.91)
- ZeroMamba: Exploring Visual State Space Model for Zero-Shot Learning (AAAI 2025 · similarity 0.91)
- ZeroDiff: Solidified Visual-semantic Correlation in Zero-Shot Learning (ICLR 2025 · similarity 0.90)
- Data-Free Generalized Zero-Shot Learning (AAAI 2024 · similarity 0.90)
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.
