AI Paper Atlas

Research map / Video / Action recognition

Towards Generalizing Temporal Action Segmentation to Unseen Views

Emad Bahrami, Olga Zatsarynna, Gianpiero Francesca, Jüergen Gall

IJCV 2026 · Video › Action recognition · approach: Temporal action segmentation · 1 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 sequence loss and action loss together (ablation)Baseline TAS model (MViTv2) · Assembly101 unseen ego viewsF1@50: 6.3 -> 9.7Improved
add sequence loss and action loss together (ablation)Baseline TAS model (MViTv2) · Assembly101 unseen exo viewsF1@50: 11.7 -> 13.2Improved
add sequence loss and action loss together (ablation)Baseline TAS model (MViTv2) · Assembly101 seen viewsF1@50: 31.1 -> 32.1Improved
add sequence loss alone (ablation)Baseline TAS model (MViTv2) · Assembly101 unseen ego viewsF1@50: 6.3 -> 8.1Improved
add action loss alone (ablation)Baseline TAS model (MViTv2) · Assembly101 unseen ego viewsF1@50: 6.3 -> 8.5Improved
include same-view action segments in action loss (ablation)Ours · Assembly101 unseen ego viewsF1@50: 9.7 -> 7.0Hurt
add adversarial view-classification loss (ablation)Ours · Assembly101 unseen ego viewsF1@50: 9.7 -> 6.4Hurt
use cosine similarity instead of MSE (ablation)Ours · Assembly101 unseen ego viewsF1@50: 7.2 -> 9.7Improved
use cosine similarity instead of KL divergence (ablation)Ours · Assembly101 unseen ego viewsF1@50: 6.5 -> 9.7Improved
use fine instead of coarse action segments in action loss (ablation)Ours · Assembly101 unseen ego viewsF1@50: 6.5 -> 9.7Improved
desynchronize views by random shift of 1 frame (ablation)Ours · Assembly101 unseen ego viewsF1@50: 9.7 -> 7.7Hurt
desynchronize views by random shift of 10 frames (ablation)Ours · Assembly101 unseen ego viewsF1@50: 9.7 -> 6.3Hurt

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