AI Paper Atlas

Research map / Video / Action recognition

Leaving Some Stones Unturned: Dynamic Feature Prioritization for Activity Detection in Streaming Video

Yu-Chuan Su, Kristen Grauman

ECCV 2016 · Video › Action recognition · approach: Disentangled action recognition · 30 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
learn dynamic feature-prioritization policy instead of random (Passive) selectionStreaming activity recognition · UCF-101 and ADL streamingcost to reach accuracy: reduces cost by 80% and 50% on UCF and ADLImproved
use instance-dependent feature selection instead of DT-Top static top featuresStreaming recognition · UCF-101 multi-classaccuracyImproved
add skip action and dynamic policy with CNN featuresStreaming recognition with CNN · UCF-101frames processed: skips 80% of the frames, achieves over 90% of the ultimate accuracyImproved
dynamic policy instead of Passive with CNN featuresStreaming recognition with CNN · UCF-101frames needed: less than half of the framesImproved
dynamic policy for batch recognitionBatch recognition · ADLfeatures extracted for same accuracy: 30% of the featuresImproved
dynamic policy instead of Object-PreferenceBatch recognition · UCF-101 and ADLspeed: 2.5 times fasterImproved
dynamic policy vs DT-Top in untrimmed binary detectionUntrimmed streaming detection · UCF-101F1No clear effect
dynamic policy vs baselines in untrimmed detectionUntrimmed streaming detection · ADLF1: nearly twice the accuracy of all baselines until 16 fpsImproved
dynamic policy with skip action at high detector speedUntrimmed streaming detection · UCF-101 at 64 fpsdetections run: 50% fewer detectionsImproved
dynamic policy instead of MMED early detectorUntrimmed streaming detection · ADL and UCF AMOCtimeliness and costImproved
train decision tree on same features as classifier f without subvolumes (ablation)Decision tree baseline · Batch recognitionaccuracyHurt
add dense trajectories to CNN featuresStreaming recognition · UCF-101accuracyImproved

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