AI Paper Atlas

Research map / Video / Action recognition

Efficient Activity Detection in Untrimmed Video with Max-Subgraph Search

Chao-Yeh Chen, Kristen Lorraine Grauman

TPAMI 2016 · Video › Action recognition · approach: Fully supervised temporal action detection · 16 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 temporal jumps (T-Jump) to T-Subgraph (ablation)T-Subgraph · UCF Sportsmean overlap accuracy: Diving 0.8106 -> 0.9091; Riding 0.5349 -> 0.3888Mixed
add temporal jumps (T-Jump) to T-Subgraph (ablation)T-Subgraph · Hollywood uncroppedmean overlap accuracy: AnswerPhone 0.3968 -> 0.3994; GetOutCar 0.2276 -> 0.2921Improved
use T-Subgraph search instead of exhaustive T-SlidingT-Subgraph · UCF Sportsdetection time (ms): 1.25x10^5 -> 1.02x10^2Improved
use T-Subgraph search instead of exhaustive T-SlidingT-Subgraph · Hollywood uncroppeddetection time (ms): 3.71x10^3 -> 6.63x10Improved
use T-Subgraph instead of branch-and-bound ST-Cube-SubvolumeT-Subgraph · Hollywood uncroppeddetection time (ms): 1.70x10^5 -> 6.63x10Improved
add temporal jumps (T-Jump), more complex graph (ablation)T-Subgraph · Hollywood uncroppeddetection time (ms): 6.63x10 -> 5.69x10^2Hurt
detect activity first (T-Subgraph output) before recognition instead of raw uncropped clipsrecognition pipeline · Hollywoodrecognition accuracy: 24.83% -> 29.66%Improved
replace HoG/HoF low-level features with object-based high-level descriptors (ablation)T-Subgraph · Hollywood uncroppedmean overlap accuracy: HugPerson 0.3869 -> 0.5292; AnswerPhone 0.3968 -> 0.1741Mixed
use T-Subgraph vs T-Sliding for multiple instancesT-Subgraph · THUMOS 2014mAP: 0.1983 -> 0.2143Improved
add temporal jumps (T-Jump) to T-Subgraph (ablation)T-Subgraph · THUMOS 2014mAP: 0.2143 -> 0.1546Hurt
add temporal jumps (T-Jump) to T-Subgraph (ablation)T-Subgraph · THUMOS 2014overlap: 0.2186 -> 0.2636Improved
add spatial links (ST-Subgraph) to temporal subgraph (ablation)T-Subgraph · MSR Action temporal overlapmean temporal overlap accuracy: Boxing 0.0541 -> 0.0989; Clapping 0.0982 -> 0.1754Improved

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