AI Paper Atlas

Research map / Video / Action recognition

Event-Guided Procedure Planning from Instructional Videos with Text Supervision

An-Lan Wang, Kun-Yu Lin, Jia-Run Du, Jingke Meng, Wei-Shi Zheng

ICCV 2023 · Video › Action recognition · approach: Action anticipation and latent actions · 18 citations (OpenAlex)

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 changedWhereMeasuredEffect
add Event-aware Prompt Generator (ablation)E3P · CrossTask (T=3)Success Rate: 23.55 -> 25.62Improved
add Action Relation Mining module (ablation)E3P · CrossTask (T=4)Success Rate: 14.85 -> 16.49Improved
add prompting-based feature extractor to the action classifier baseline (ablation)baseline · CrossTask (T=3)Success Rate: 22.56 -> 23.55Improved
add Event-aware Prompt Generator to full model without it (ablation)E3P · CrossTask (T=4)Success Rate: 14.23 -> 16.49Improved
add DropRelation regularization (drop rate 20%) to ARM (ablation)E3P · CrossTask (T=4)Success Rate: 15.69 -> 16.49Improved
increase DropRelation rate beyond optimum (40%) (ablation)E3P · CrossTask (T=3)Success Rate: 26.40 -> 25.76Hurt
use pre-trained CLIP text encoder with hand-crafted prompts (ablation)E3P · CrossTask (T=4)Success Rate: 14.70 -> 16.49Improved
add ARM to reduce action transition matrix error (ablation)E3P · CrossTask (T=4)mean Absolute Error: 3.21 -> 2.55Improved
event-guided paradigm instead of direct state-to-action predictiontext-supervised procedure planning · COIN (T=3)Success Rate: 15.40 -> 19.57Improved
E3P vs P3IV on mIoU for CrossTask T=3E3P · CrossTask (T=3)mIoU: 73.89 -> 74.05No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy