AI Paper Atlas

Research map / Video / Action recognition

StepFormer: Self-Supervised Step Discovery and Localization in Instructional Videos

Nikita Dvornik, Isma Hadji, Ran Zhang, Konstantinos G. Derpanis, Richard P. Wildes, Allan D. Jepson

CVPR 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
Train on subtitles and transfer without dataset-specific fine-tuningStepFormer · CrossTask, ProceL, and COINF1Improved
Add sequence-to-sequence alignment loss (ablation)StepFormer · CrossTask unsupervised segmentationF1: 17.3 -> 28.3Improved
Add sequence-to-sequence alignment loss (ablation)StepFormer · CrossTask zero-shot localizationIoU: 4.6 -> 23.7Improved
Add global contrastive loss (ablation)StepFormer · CrossTask unsupervised segmentationF1: 20.9 -> 28.3Improved
Add global contrastive loss (ablation)StepFormer · CrossTask zero-shot localizationIoU: 11.5 -> 23.7Improved
Add diversity regularizer (ablation)StepFormer · CrossTask unsupervised segmentationF1: 26.7 -> 28.3Improved
Add diversity regularizer (ablation)StepFormer · CrossTask zero-shot localizationIoU: 16.9 -> 23.7Improved
Add smoothness regularizer (ablation)StepFormer · CrossTask unsupervised segmentationF1: 27.1 -> 28.3Improved
Add smoothness regularizer (ablation)StepFormer · CrossTask zero-shot localizationIoU: 17.7 -> 23.7Improved
Use ordered Drop-DTW inference instead of order-agnostic frame-to-slot assignment (ablation)StepFormer · CrossTask unsupervised segmentationF1: 28.1 -> 28.3No clear effect
Use ordered Drop-DTW inference instead of order-agnostic frame-to-slot assignment (ablation)StepFormer · CrossTask zero-shot localizationIoU: 20.1 -> 23.7Improved
Use 48 rather than 32 step slots (ablation)StepFormer · CrossTask unsupervised segmentationF1: 28.3 -> 26.5Hurt

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