Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning
A Hybrid Egocentric Activity Anticipation Framework via Memory-Augmented Recurrent and One-Shot Representation Forecasting
Tianshan Liu, Kin-Man Lam
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 |
|---|---|---|---|
| use HRO instead of the strongest listed baseline AVT | EPIC-Kitchens activity anticipation · EPIC-Kitchens validation set, 1s | Top-5 activity accuracy (%): 37.60 -> 37.42 | No clear effect |
| use HRO instead of SRL | EGTEA Gaze+ activity anticipation · EGTEA Gaze+, 1s | Top-5 activity accuracy (%): 70.67 -> 71.46 | Improved |
| add an external memory bank to vanilla GRU representation forecasting (ablation) | HRO · EPIC-Kitchens validation set | Top-5 activity accuracy (%) at 1s: 34.81 -> 35.18 | Improved |
| add sparsity regularization to memory-bank learning (ablation) | HRO with a memory bank · EPIC-Kitchens validation set | Top-5 activity accuracy (%) at 1s: 35.18 -> 35.36 | Improved |
| add diversity regularization to memory-bank learning (ablation) | HRO with a memory bank · EPIC-Kitchens validation set | Top-5 activity accuracy (%) at 1s: 35.18 -> 35.52 | Improved |
| add both diversity and sparsity losses to memory-bank learning (ablation) | HRO · EPIC-Kitchens validation set | Top-5 activity accuracy (%) at 1s: 34.81 -> 35.64 | Improved |
| increase memory-bank size from small to larger sizes (ablation) | HRO · EPIC-Kitchens validation set | Top-5 activity accuracy (%) at 1s | Mixed |
| add future-semantics exploration using recurrent forecasting (ablation) | HRO · EPIC-Kitchens validation set | Top-5 activity accuracy (%) at 1s: 35.64 -> 36.53 | Improved |
| replace recurrent future-semantics forecasting with one-shot transfer (ablation) | HRO · EPIC-Kitchens validation set | Top-5 activity accuracy (%) at 1s: 36.53 -> 37.42 | Improved |
Closest papers by meaning
- Self-Regulated Learning for Egocentric Video Activity Anticipation (TPAMI 2021 · similarity 0.91)
- Rethinking Learning Approaches for Long-Term Action Anticipation (ECCV 2022 · similarity 0.88)
- Open-Vocabulary Long Term Action Anticipation (ECCV 2026 · similarity 0.88)
- Semantically Guided Representation Learning For Action Anticipation (ECCV 2024 · similarity 0.88)
- Time-Conditioned Action Anticipation in One Shot (CVPR 2019 · similarity 0.88)
- AntGPT: Can Large Language Models Help Long-term Action Anticipation from Videos? (ICLR 2024 · similarity 0.85)
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.
