AI Paper Atlas

Research map / Video / Video-language understanding

Measuring the Importance of Temporal Features in Video Saliency

Matthias Tangemann, Matthias Kümmerer, Thomas S.A. Wallis, Matthias Bethge

ECCV 2020 · Video › Video-language understanding · approach: Event and anomaly reasoning in video · 13 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
average VGG features over 16 recent frames (DeepGaze MR) instead of frame-by-frame predictionDeepGaze II · LEDOV valinformation gain (bit): 1.216 -> 1.445Improved
average VGG features over 16 recent frames (DeepGaze MR) instead of frame-by-frame predictionDeepGaze II · LEDOV testinformation gain (bit): 1.117 -> 1.367Improved
average VGG features over 16 recent frames (DeepGaze MR) instead of frame-by-frame predictionDeepGaze II · DIEMinformation gain (bit): 0.674 -> 0.660Hurt
average VGG features over 16 recent frames (DeepGaze MR) instead of frame-by-frame predictionDeepGaze II · Meta-benchmark (LEDOV & DIEM)information gain (bit): 0.787 -> 0.528Hurt
static model DeepGaze MR versus temporal video model TASED-NetDeepGaze MR · LEDOV valNSS: 3.498 -> 3.857Improved
static model DeepGaze MR versus temporal video model TASED-NetDeepGaze MR · DIEMNSS: 3.194 -> 3.116Hurt
static model DeepGaze MR versus temporal video model TASED-NetDeepGaze MR · Meta-benchmark (LEDOV & DIEM)NSS: 2.855 -> 2.458Hurt
static features explain share of explainable information on LEDOVDeepGaze MR · LEDOV valpercent explainable information: 62.8 -> 74.6Improved
select window length 16 frames by grid search (ablation)DeepGaze MR · LEDOVinformation gainImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 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