AI Paper Atlas

Research map / Video / Action recognition

ASPnet: Action Segmentation With Shared-Private Representation of Multiple Data Sources

Beatrice van Amsterdam, Abdolrahim Kadkhodamohammadi, Imanol Luengo, Danail Stoyanov

CVPR 2023 · Video › Action recognition · approach: Temporal action segmentation · 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
Fuse video and accelerometer features with shared-private representations and a temporal attention bottleneckASFormer · 50SaladsF1@10: 84.8 -> 89.2Improved
Add accelerometer input to Br-Prompt RGB and optical flowBr-Prompt+ASPnet · 50SaladsF1@10: 91.2 -> 92.7Improved
Use ASPnet with RGB-flow inputs instead of ASFormer feature concatenationASFormer · 50SaladsF1@10: 87.8 -> 88.6Improved
Use ASPnet instead of ASFormer with the same video input featuresASFormer · BreakfastF1@10: 76.0 -> 78.1Improved
Use ASPnet instead of MA-TCN for video-based segmentationMA-TCN · RARP45F1@10: 83.7 -> 86.7Improved
Add moving-average post-processing (ablation)ASPnet · 50SaladsF1@10: 89.2 -> 90.4Improved
Initialize the attention bottleneck with learned shared features rather than a Gaussian (ablation)ASPnet · 50SaladsF1@10: 86.5 -> 89.2Improved
Use MMD instead of Euclidean distance to align shared features (ablation)ASPnet · 50SaladsF1@10: 87.0 -> 89.2Improved
Add private features to the prediction by removing the private-feature mask (ablation)ASPnet · 50SaladsF1@10: 83.0 -> 90.4Improved
Add shared bottleneck features to the prediction by removing the shared-feature mask (ablation)ASPnet · 50SaladsF1@10: 86.5 -> 90.4Improved
Share weights across encoder streams (ablation)ASPnet · 50SaladsF1@10: 91.3 -> 91.0No clear effect
Use scalable MMD against the average bottleneck with shared encoder weights (ablation)ASPnet · 50SaladsF1@10: 91.0 -> 90.9No clear effect

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