AI Paper Atlas

Research map / Video / Video segmentation and tracking

A Sparse Sampling-Based Framework for Semantic Fast-Forward of First-Person Videos

Michel M. Silva, Washington Ramos, Mário F. M. Campos, Erickson R. Nascimento

TPAMI 2020 · Video › Video segmentation and tracking · approach: Self-supervised video representation learning · 10 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 Fill Gap Between Segments step (vs SAS) (ablation)SAS · ASDdiscontinuity RMSE: 23.0 -> 10.1Improved
add gap filling and speed-up smoothing (vs SAS) (ablation)SAS · ASDmean squared speed-up difference: 39.6 -> 18.7Improved
add gap filling and speed-up smoothing (vs SAS) (ablation)SAS · DoMSEVmean squared speed-up difference: 76.9 -> 23.0Improved
add gap filling and transition smoothing (vs SAS) (ablation)SAS · DoMSEV (all classes mean)semantic retained (%): 32.3 -> 33.1Improved
add gap filling and transition smoothing (vs SAS) (ablation)SAS · DoMSEV (all classes mean)instability: 33.5 -> 34.6Hurt
add gap filling and transition smoothing (vs SAS) (ablation)SAS · DoMSEV (all classes mean)discontinuity: 31.7 -> 12.4Improved
use LLC instead of Lasso sparse coding for frame sampling (ablation)frame sampling · ASD sparse coding ablationrunning time (s): 21.3 -> 1.1Improved
use LLC instead of OMP for frame sampling (ablation)frame sampling · ASD sparse coding ablationrunning time (s): 27.5 -> 1.1Improved
use LLC instead of OMP for frame sampling (ablation)frame sampling · ASD sparse coding ablationdiscontinuity: 5.9 -> 21.9Hurt
use LLC instead of Lasso for frame sampling (ablation)frame sampling · ASD sparse coding ablationinstability: 36.0 -> 34.4Improved
use LLC instead of OMP for frame sampling (ablation)frame sampling · ASD sparse coding ablationsemantic retained (%): 26.8 -> 25.6Hurt
replace handcrafted 446d features with ResNet152 deep features (ablation)sparse frame sampling · ASD features ablation (mean)semantic retained (%): 27.1 -> 28.6Improved

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