Research map / Video / Action recognition
A Perceptual Prediction Framework for Self Supervised Event Segmentation
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 self-supervised perceptual prediction for event segmentation | LSTM + AL · Breakfast Actions | MoF: 34.6 -> 42.9 | Helped |
| use self-supervised perceptual prediction for event segmentation | LSTM + AL · 50 Salads (Eval) | MoF: 54.0% -> 60.6% | Helped |
| use self-supervised perceptual prediction for event segmentation | LSTM + AL · INRIA Instructional Videos | F1: 32.2% -> 39.7 | Helped |
| add adaptive learning (ablation) | RNN predictor · INRIA Instructional Videos | F1: 25.9 -> 29.4 | Helped |
| add adaptive learning (ablation) | LSTM predictor · INRIA Instructional Videos | F1: 36.4 -> 39.7 | Helped |
| use LSTM instead of RNN predictor (ablation) | predictive event segmentation with adaptive learning · INRIA Instructional Videos | F1: 29.4 -> 39.7 | Helped |
| pretrain with self-supervised predictive features | VGG16 + LSTM activity recognition · Breakfast Actions | Precision: 33.54% -> 37.87% | Helped |
| increase prediction history window (ablation) | perceptual prediction segmentation · unspecified ablation setting | IoU | Helped |
| increase prediction history window (ablation) | perceptual prediction segmentation · unspecified ablation setting | MoF | Hurt |
| lower the prediction error threshold (ablation) | perceptual prediction segmentation · unspecified ablation setting | over-segmentation | Hurt |
Closest papers by meaning
- Temporal Action Segmentation From Timestamp Supervision (CVPR 2021 · similarity 0.89)
- Action Sets: Weakly Supervised Action Segmentation Without Ordering Constraints (CVPR 2018 · similarity 0.89)
- SCT: Set Constrained Temporal Transformer for Set Supervised Action Segmentation (CVPR 2020 · similarity 0.88)
- No More Shortcuts: Realizing the Potential of Temporal Self-Supervision (AAAI 2024 · similarity 0.88)
- Action Segmentation With Joint Self-Supervised Temporal Domain Adaptation (CVPR 2020 · similarity 0.88)
- OnlineTAS: An Online Baseline for Temporal Action Segmentation (NeurIPS 2024 · 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.