AI Paper Atlas

Research map / Video / Action recognition

StrokeRehab: A Benchmark Dataset for Sub-second Action Identification

Aakash Kaku, Kangning Liu, Avinash Parnandi, Haresh Rengaraj Rajamohan, Kannan Venkataramanan, Anita Venkatesan, Audre Wirtanen, Natasha Pandit, Heidi Schambra, Carlos Fernandez-Granda

NeurIPS 2022 · Video › Action recognition · approach: Disentangled action recognition · 5 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
replace MS-TCN segmentation with Seg2seqMS-TCN · StrokeRehab stroke-patient test, videoaction error rate: 0.392 -> 0.322Improved
replace MS-TCN with Raw2seqMS-TCN · StrokeRehab stroke-patient test, sensoraction error rate: 0.330 -> 0.305Improved
use Raw2seq instead of Seg2seq on sensor dataSeg2seq · StrokeRehab stroke-patient test, sensoraction error rate: 0.337 -> 0.305Improved
add smoothing window to MS-TCN (ablation)MS-TCN · StrokeRehab stroke-patient test, videoaction error rate: 0.392 -> 0.390No clear effect
add boundary-detection refinement (ASRF over MS-TCN) (ablation)MS-TCN · StrokeRehab stroke-patient test, videoaction error rate: 0.392 -> 0.436Hurt
add smoothing window to MS-TCN (ablation)MS-TCN · 50Saladsaction error rate: 0.43 -> 0.32Improved
add smoothing window to MS-TCN (ablation)MS-TCN · Breakfastaction error rate: 0.97 -> 0.51Improved
replace ASRF with Seg2seqASRF · Jigsawsaction error rate: 0.31 -> 0.17Improved
replace ASRF with Raw2seqASRF · Breakfastaction error rate: 0.45 -> 0.55Hurt
select model by validation AER instead of frame-wise accuracy (ablation)MS-TCN · StrokeRehab stroke-patient test, videoaction error rate: 0.408 -> 0.392Improved
train Raw2seq on stroke patients instead of healthy subjects (ablation)Raw2seq · StrokeRehab stroke-patient testaction error rate: 0.405 -> 0.305Improved
train Raw2seq on stroke patients instead of healthy subjects (ablation)Raw2seq · StrokeRehab severe-impairment testaction error rate: 0.819 -> 0.612Improved

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